ENHANCED VIRTUAL ROUTER IN NETWORK TO REDUCE

PACKET DISCARD CAUSED BY THERMAL ENERGY

1. INTRODUCTION

1.1 Preamble

Heat dissipation in electronics devices is one of the major problems in science &
technologies. Many researchers analyzed that to minimize the heat dissipation but still
remain as the stable problem in electronic devices. The amount of heat output in electronic
device is equal to the power input, if there are no other energy interactions. There are
several techniques for cooling including various styles of heat sinks, thermoelectric
coolers, forced air systems fans, heat pipes, and others. This kind of cooling devices are
based on electronic circuits that are used in antenna, switching devices etc. Apart from this
thermal consideration, still data communication network industry are facing a
slowdown. The problem of Internet traffic affects the usage of mobile Internet, e-
commerce, content rich applications, and multi-media services. A recent study shows that
Internet traffic has increased by approximately 80% annually in the past three years which
in turn leads to increase in heat dissipation simultaneously. This research is broadly related
to the areas of data communication, wired and wireless network management, with
special focus on avoiding packet discarding techniques in buffer unit of the router,
(switching device in network) in order to enhance the quality of service in internet.
By enhancing the quality of electronic components, sustainable quality in data
communication network can be maintained. At the same time, network power requirements
will scale with capacity growth. Keeping in mind, the traffic demands will soon reach tens
of Tbps, (Tera bits per second) power demands may limit or decelerate network growth.
This proposed research plays an important role in networking aspect and from power
density perspective. Heat dissipation in electronic devices which is a major problem in data
communication network disturb the performance of the network. There are many existing
technologies that can be used to address the problems in thermal management for data
communication network applications. Internetworking consists of several nodes that
perform three important functions strongly participate in power consumption with the
distribution frequency as processing (22%), switching/routing (34%), and electrical

1
 regeneration (27%).
Switching and Routing is the basic function of nodes. It dissipates 34% of thermal
energy. During this function, thermal variations occur in these nodes which affects entire
throughput of Multimedia communication network. In order to overcome this problem, it is
internally rebuilding router unit by using special silicon oil sheets to reduce thermal power
called upgraded virtual router. After placing silicon sheets in buffer unit exclusively, heat
dissipation and its packet drop in the network is traced out using packet tracer software. The
upgraded version of router in buffer unit reduces thermal leakages 5-10 % of existing router
and it increases throughput of system by 94%. And also congestion and collision of the
network is reduced by 0.0074% in all nodes.
The proposed research deals to improve the power demands in communication
networks. More importantly, it deals with arising challenges in that field, particularly if the
network growth continues at the current pace.
World Wide Web (Audio and video information) Multimedia messages have the
enormous number of packets to forward from node to node. Buffer plays a vital role to store
and forward the packets without discarding. Hence the main objective of this research is
focused on the buffer Design unit in order to enhance the quality of service in multimedia
messages. The other objectives of this research have been
1. To analyse each node in network and to find heat dissipation causes in packet discard
mechanism.
2. To identify the concepts of improving the efficiency and effectiveness of Router in
order to achieve congestion and collision free network.
3. To design buffer unit to avoid packet discarding function due to thermal leakage by
using oil silicon sheets and path diversity.
4. To prove the above sequence by packet tracer software using enhanced hardware
structure of router.

1.2 DATA COMMUNICATION

Data communications (DC) is the process of computing and communication

technologies to transfer data (Multimedia msgs, Text, Character, Symbol) from one place to
another, and vice versa. The important elements to transfer the data’s are 1.Sender 2.Receiver
3.Protocol 4.message 5.Medium. It enables the movement of electronic or digital data
between two or more nodes.

2
 For data communication, the communicating devices must be the part of a
communication system which made up of a combination of hardware and software. The
effectiveness of a data communication system depends on the following fundamental
characteristics where present in hardware and software.

1. Delivery: The System must deliver data to the correct destination. Data must be arrived by
the intended device or user and only by that device.

2. Accuracy: The system must deliver data accurately. Data that have been altered in
transmission and left uncorrected are rustles. And also data’s never to be lost. If packets are
lost due to any circumstance the communication becomes great flaw in Multimedia messages.
Hence the communication becomes endless communication.

3. Timeliness: The system must deliver data in a timely manner.

In order to have full-fledged communication the following network is the basic function
nodes where present in data communication.

Figure 1

3
If the messages passing far nodes it needs to forward the information from path to path
routing algorithm

Figure 2

In the above mentioned diagram 4 nodes are present in the communication. In Data
communications networks, is either a redistribution point (RP) or a communication endpoint.
The definition of a node depends on the network and protocol layer (PL) referred to. A
physical network node is the assembly of an active electronic device (AED) and passive
distribution point. An active device is attached to a network, and is capable of creating,
receiving, or transmitting information over a communications channel. A passive distribution
point (PDP) such as a distribution frame or patch panel is consequently not a node.

1.3 COMPUTER NETWORK

In data communication, a physical network node may either be data communication

equipment (DCE) such as a modem, hub, bridge, Router, switch, data terminal
equipment (DTE).

If the network in question is the Internet or an Intranet, many physical network nodes
are host computers, also known as Internet nodes, identified by an IP address and all hosts are
physical network nodes. However, some data link layer devices such as switches, bridges
and wireless access points do not have an IP host address and are not considered to be
Internet nodes or hosts, but as physical network nodes and LAN nodes. In the data
communication active nodes are made up of electronic devices. Hence it is the major role of
delivering messages from one node to another node.

By enhancing the quality of electronic components, sustainable quality in data

4
communication network can be maintained. At the same time, network power requirements
will scale with capacity growth. Keeping in mind, the traffic demands will soon reach tens
of Tbps, (Tera bits per second) power demands may limit or decelerate network growth.
This research work plays an important role in networking aspect and from a power density
perspective also. Heat dissipation in electronic devices which is a major problem in data
communication network disturbs the performance of the network. There are many existing
technologies that can be used to address the problems in thermal management for data
communication network applications.
Networking devices may include gateways, routers, network
bridges, modems, wireless access points, networking cables, line drivers, switches, hubs,
and repeaters; and may also include hybrid network devices such as multilayer
switches, protocol converters, bridge routers, proxy servers, firewalls, network address
translators, multiplexers, network interface controllers, wireless network interface
controllers, ISDN terminal adapters and other related hardware.

1.3.1 NETWOKING POWER DISSIPATION

Internetworking consists of several nodes that perform three important functions
strongly participate in power consumption with the distribution frequency as: processing
(22%), switching/routing (34%), and electrical regeneration (27%).

22%
27%
Processing
Switching /Routing
Electrical Generation

34%

Figure 3

5
 The above figure 3 active nodes are radiating heat energy in the specified percentage
83%, where as passive nodes are losing, 17% of energy in various way. Now this research
work focusing the heat dissipation in the active node of Router Unit (RU).

Basic Parameters to enhance quality of service in each layer

Table 1

Name of the Layer QOS Parameters

Application Layer Data consistency, Detection Chance, Data originality, Reply time
and System life time

Transport Layer Rate, Delay, Medium and reliability

Network Layer Power efficiency, Routing Strength, Bottleneck chance, Routing

Protection and path delay

MAC layer Communication reliability, Throughput and range of

communication

Physical Layer Communication, Data Processing and sensing Components

1.3.2 METHODS OF HEAT DISSIPATION

Electronic devices are made up of collection of Semiconductors material which

generates heat during use. Conduction, Convection, Radiation are the three basic methods
why semiconductor materials are dissipating heat energy.

Heat transfer refers to the movement of heat from one point to the other due to temperature
difference. There are three main mechanisms by which heat can be transferred from one
region to another:

Conduction is a method of heat transfer in which heat is exchanged between two elements
that are situated close to one another or in still fluids or gaseous materials.

Convection refers to heat transfer in a liquid or gas by the circulation or flow of the liquid or
gas from one region to another.

Radiation refers to heat exchange through electromagnetic heat exchange in the form of
waves and rays. There is no material connection between the two regions involved. This heat

6
must be removed in order to prevent further heating and consequent destruction of the
semiconductor.

Due to this convection and Radiation heat transferring occurs in semiconductor devices. This
heat dissipation in electronics devices is the greatest problem in modern science of
technologies. There are so many research work had taken place to minimize the heat
dissipation hence this is stable problem in electronic devices. Heat dissipation problem
broadly affecting wired and wireless communication networking where huge number of
electronic devices is participating to transmit the multimedia messages from one node to
another node.
The amount of heat output in electronic device is equal to the power input, if there are
no other energy interactions. There are several techniques for cooling including various styles
of heat sinks, thermoelectric coolers, forced air systems and fans, heat pipes, and others. This
kind cooling based of electronic circuitry is placed into communication field like antenna,
switching devices etc. Apart from this thermal consideration, still communication fields is
facing slowdown progress and this affects communication industry like Internet traffic
continues to grow rapidly, mobile Internet, e-commerce, content rich applications, and multi-
media services. By the means of results, Internet traffic has increased by approximately 80%
yearly in each of the last three years, and will continue to grow at the heat dissipation in the
circuitry too.

Structure and Connection Technology

For heat transport within semiconductor elements, only heat transfer through conduction
comes into consideration. Temperatures do not rise too high during the operation of a
semiconductor element; the effects of structure and connection technology on heat dissipation
are of particular importance. For this reason, the thermal conductivity and coefficient of
thermal expansion of the materials used are crucial. Rules pertaining to high reliability must
always be adhered to.

If the layered surfaces of a plate with thickness d are at temperatures T1 and T2, the amount
of heat Q that flows through the surface during time t is proportional to thermal gradient

7
 Figure 4
Conduction

When the elements of devices in direct contact heat transfers by means of energy to
Conduction. Adjacent atoms of higher energy vibrate against one another, which transfers the
higher energy to the lower energy, or higher temperature. Atoms of higher intensity and
higher heat will vibrate, thereby moving the electrons to areas of lower intensity and lower
heat. Fluids and gases are less conductive than solids due to the fact they are less dense,
meaning that there is a larger distance between atoms.

Convection

Convection describes heat transfer between a surface and a liquid or gas in motion. As the
fluid or gas travels faster, the convective heat transfers increases. Two types of convection
are natural convection and forced convection. In forced convection the fluid is forced to
travel over the surface by a fan or pump or some other external source.

Radiation

Radiation refers to the transfer of heat through empty space. This form of heat transfer occurs
without intervening medium. When the molecules are radiating collection of atoms are
moving upwards which creates heat dissipation. Here in the nodes this radiation phenomena
affected and information are lost.

1.3.2 NECESSITY OF THERMAL MANAGEMENT

Thermal Management is the ability to control the temperature of a device which may be in
computer network it is meant of Technology based on Thermodynamics and Heat Transfer.
Thermal Management is therefore describing all possible means and processes like heat
transfer, conduction, convection, condensation and radiation, etc. to increase or decrease the
temperature and subsequently reducing packet loss mechanism in routers

8
Ability of the Routers is totally decreasing when system transfers the energy, work and mass.
The content will define the ability of the system to use these transfers to change the systems
temperature and temperature distribution by storing and discharging energy, work and mass.
Reducing the work and mass of the device, causes the reduction of heat dissipation. In
Routers systems it is necessary to define "sub-systems" to describe the interaction between
important parts of the each block and main system. Before the thermal treatment, Routers
heat dissipation platforms percentage is very important. Hence thermal simulations, in
conventional designs, the routers on the top layers frequently to avoid thermal emergency.
Therefore, the proposed architecture removes all routers on the top layers. At 90°C, our
architecture receives 1.4 times as many packets when compared to conventional designs. If
the temperature constraint is set at 80°C, our architecture can receive 2 times as many
packets. In addition, this new architecture is energy-efficient because the average number of
nodes is reduced. Higher energy levels of the materials in the system are causing the
temperature of these materials to rise.

Temperature is in Thermal Management the main state variable of a router shows the energy
level of the system. Each system has a given ability to store or dissolve energy. This ability is
called the "heat capacity" of the Router, and it can be calculated by the mass and the specific
heat capacity of all materials in the system. When energy is transferred into the router or
energy is created by an internal source in the system, the energy level of the system can rise.
The specific heat capacity defines the connection between absorbed or dissolved energy and
the change of temperature.

When the router specifications do not fulfill the equipment requirements under the ambient
conditions, this deviation should be described in quality and quantity. Thermal Management
concepts use different strategies to solve typical gaps between requirements and
specifications. Basic concepts are heating, cooling, and heat removing cycling. Principles
used for these concepts enclose enhancement of heat transfer thermal insulation, selective
heat transfer and active cooling with Thermoelectric.

Currently multiprocessors scale to increasing numbers of cores.. For many high-performance

computing applications, the bandwidth and frequency communications can play a vital role in
efficient implementation due to the use of data-parallel or data-centric algorithms. Electronic
interconnected systems are increasingly bound by their communications infrastructure and

9
the associated with power dissipation of high-bandwidth multimedia messages movement.
Recent advances in chip-scale silicon technologies have created the potential for developing
interconnection networks that can offer highly energy efficient communications and
significantly improve computing performance-per-Watt.

1.3.4 POWER EFFICIENCY

Power consumption of Internet infrastructure and related information and communications

technology (ICT) infrastructure is based on equipment inventory. Traditional sales data of
telecommunications equipment, together with information about the energy consumption
provide a good knowledge of power efficiency and its working methods. It is not expose the
inter-play of packet consequential movement’s but only exercise about the growth of the
Tbps. This is important for estimating how future growth trends may change power
consumption patterns as more Internet-based services are in QOS taken up.

Energy consumption of power -saving switch wireless router, ordinary router and energy-
saving switch router's power were tested. Ordinary router’s average power is stably around
4.25 w, the average power of energy-saving switch router is stably around 5.85 W, but in the
same environment the running time of the former is more than 2 times to the latter. A
wireless router will save power 13.359 kw/h one year which depends on the wireless router in
experiment.

Figure 5
10
 We estimate the worldwide network Power consumption PW by extrapolating the electricity
consumption of the representative sample. Pi = Network electricity consumption of operator
The calculation of PW is performed for each year in 2013-2017. For 2017, we estimate the
worldwide electricity consumption by extrapolating the values of the previous years.

MW=Maximum power

MS=Maximum subscriber Power

PS=Sample Power

PW = MW/ MS(w) ×PS(w)

∑ i=1 (wi × pi)

Cooling overhead factors and its worldwide electricity use per type of equipment Power
Demand estimates are adaptations of the values

Table 2

Equipment Cooling overhead Power Power

2013 (TWh) 2017 (TWh)
Switching - 10/100 1.38 10.7 15.7

Switching- 10/100/1000 1.38 17.5 19.5

Routers -Small & Medium 1.75 4.2 8.2

Enterprise WLAN 1.00 2.3 5.3

Security-Small & Medium 1.75 7.7 10.7

Total 42.4 59.4

Following graph shows that the percentage variation of Power consumption in switching
and routing in 2013 and 2017 .

11
 25

19.5
20
17.5
15.7
15
Cooling overhead
10.7 10.7
10 8.2 Power 2013 (TWh)
7.7
5.3 Power 2017 (TWh)
4.2
5
1.75 2.3 1.75
1.38 1.38 1
0
Switching - Switching- Routers -Small & Enterprise WLAN Security-Small &
10/100 10/100/1000 Medium Medium

Figure 6

1.3.5 NETWORK PERFORMNACE

The performance of the network is the foundation to any applications and data, so
issues at any layer will lead to a bad experience. There are many reasons of lacking of
network performance because many problems reducing Quality of Service ( QOS).The
following four causes are among the most common issues currently which dropping the
packets:

i) Link Congestion

Data’s are travelling multiple devices and links during its trip across the network as in figure
2.If one of these links is at full capacity when your data arrives, it must wait its turn before
being sent across the wire which is calling as queuing. If a network device is falling very far
behind the new data never to wait or queue, so it does the clear task of discarding the
information. Data is “discarded” which causes congestion and collision of the network and
also it reduces performance of the connectivity. Real time application slows down its transfer
speed, and re-transmits the data. For example File download, an email, or another non real-
time application, the effect will be minimal as long as the packet loss doesn’t continue to
happen.

The effect of packet loss due to network congestion can be reduced by enhancing
12
  Increase the bandwidth of the congested link(s).

 Implement Quality of Service (QoS) to give priority to real-time traffic.

ii) Router/Switch/Firewall Performance

Bandwidth is adequate; still face an issue if router/switch/firewall is not able to keep up with
the traffic. The traffic is reaching the device; memory is maxed out and not able to handle
extra traffic. This results in packet loss for all traffic that is beyond the capacity of the box.
Here whenever the packet loss occurs in device, there are service points which sends
Sequence number of the lost packet were send to sender end and receiver end, which affects
the network heavily.

It can be avoided to replace the hardware with a new appliance that can handle current
maximum throughput.

iii) Software issues on a network device

When the software written for the network devices is perfect, there is less no of chances to
create problem in the network. These network devices are extremely complex, and it is a
matter of time before stumble upon a bug.

This software can cause new features to not work at all when deploy them, or can go
undetected for awhile before noticing performance issues. Once the performance issue is
detected and troubleshooting has started, these types of issues are usually found using system
logs and packet captures.

This kind of packet loss can be avoided by upgrading the software on the affected device(s).

iii). Faulty Hardware or Cabling

Hardware utilization is within specification are very important. The next common issue that
can lead to drops would be a physical component that is malfunctioning. If hardware is not

13
working properly, it will usually lead to error messages being seen on the console of the
device or within system logs.

The faulty hardware must be replaced, or the fault link must be repaired.

These are the most common for packet loss on a network, but there are many other reasons
that can contribute to packet drops. The best way to determine the root cause is through a
network assessment and detailed troubleshooting.

1.4 ROUTERS

A router is a network gateway device, it connects two or more networks and that the router
operates at the network layer of the OSI model. It contains a processor, several kinds of
digital memory, and input-output interfaces.

Router firmware

Router working mode and its memory stores an embedded operating system (OS).These
operating system are manufactured into a binary firmware which is c called commonly as
router firmware. To forward the packets into correct destination router is configured by
routing table which segregate both incoming and outgoing traffic based on the addresses of
senders and receivers.

Most powerful and largest network routers are the basic Internet backbone. These kinds of
powerful routers must manage many terabits of data flowing all the way through and between
Internet Service Provider (ISP) networks.

Essential of router networking functions:

 Sharing of Terabits of information.

 Forwarding the packets into nearer node according to the correct destination.
 Finding shortest path between source and destination.
 Segregating the packets according to the destination address route.
 Verifying the incoming packets duplication and discard it.

1.4.1 TYPES OF ROUTERS

14
Broadband Routers

Broadband routers can do different types of things. Broadband routers can be used to connect
computers or to connect to the Internet.

Wireless Routers

Wireless routers create a wireless signal in your home or office. So, any PC within range of
Wireless routers can connect it and use your Internet.

Other Type of Router

Edge Router

This type of router are placed at the edge of the ISP network, they are normally configured to
external protocol like BGP (Border gateway protocol) to another BGP of other ISP or large
organisation.

Subscriber Edge Router

This type of router belongs to an end user (enterprise) organization. It’s configured to
broadcast external BGP to it’s provider’s AS(s)

Inter-provider Border Router

This type of router is for Interconnecting ISPs, this is a BGP speaking router that maintains
BGP sessions with other BGP speaking routers.

Core Router

A router that resides within the middle or backbone of the LAN network rather than at its
periphery. Its a core router provides a step down backbone, interconnecting the distribution
routers from multiple building of a campus (LAN), or Large enterprise Location (WAN).
They tend to be optimized for a high bandwidth.

Wired and Wireless Routers

15
IP wired and wireless router. Wired and wireless router is able to maintain routing and
configuration information in their routing table. They also provide the service of filtering
traffic of incoming and outgoing packets based on IP addresses.

This is the Router which has been taken for our research and working technology is one of
the great models for forwarding the messages.

Following is the Model for the research work.

RAM Interface

RAM,ROM,Flash,
ROM CPU Interface, CPU,
Buffer

Flash BUFFER

Figure 7

ROUTING FOR RUGGED ENVIRONMENTS

The Cisco 2000 Series Connected Grid Router is designed specifically for the harsh, rugged
environments often found in the energy and utility industries. It is highly secure, reliable, and
scalable infrastructure, the Cisco 2000 Series provides industry-leading routing capabilities to
support the Smart Grid and other energy delivery communications needs.

ROUTER USED FOR RESEARCH APPROACH

The Cisco 2010 Connected Grid Router, the first in the Cisco 2000 Series, is an ideal
platform to support the communications requirements of the energy delivery infrastructure
across the generation, transmission, and distribution sectors. This product has been
extensively tested and is KEMA-certified to meet challenging substation compliance
standards, including IEEE 1613 and IEC 61850-3.

16
 Software delivers the benefits of integrated security for NERC/CIP compliance, quality of
service, and network management to help ensure integrity and priority of operational data
communications.

FEATURES AND CAPABILITIES

 Rugged industrial design and substation compliance with IEC-61850-3 and IEEE 1613 for
utility substation environments

 Integrated security to help utilities address compliance with critical infrastructure

protection mandates.

 High-availability design for optimum network up time and redundancy

 Network and device management tools for deployments, upgrades, and remote monitoring

 Advanced quality of service (QoS) capabilities to support mission-critical substation

communications such as Supervisory Control and Data Acquisition (SCADA).It is
categories into three models “b,” “g,” or “n.” It refers to the wireless communication
standard on which the router is based: 802.11b, 802.11g, and 802.11n.

Wireless communications operate in two bands—2.4GHz and 5GHz. Routers based on

802.11b and 802.11g standards and some “n” routers—use the 2.4GHz band; however,
the 802.11n standard allows wireless devices to use the 5GHz or 2.4GHz band. Routers that
operate only in the 2.4GHz band are referred to as “single band”.. Wireless “n” dual-band
routers are also available as “selectable” or “simultaneous” models, sometimes referred to as
“single radio” or “two radio,” respectively. A selectable wireless “n” router, such as the Cisco
RV220W can operate in either 2.4GHz or 5GHz mode; a simultaneous “n” device can
operate in both frequencies at the same time.

In Wi-Fi wireless networking, dual band is the capability to transmit on the 5 GHz band of
802.11a and also the 2.4 GHz band used by 802.11b, 802.11g, and 802.11n. Unlike ordinary
Wi-Fi equipment that only supports one signal band, dual-band routers contain two different
types of wireless radios that can support connections on both 2.4 GHz and 5 GHz.

17
 Figure 8

Many mobile hotspot devices only work with certain brands of cell service.

Choosing a Router

There are many different types of routers available. From least expensive to top rated, below
are some of the routers available, and they’re all available on Amazon.com:

1.4.2 BUFFER SIZE IN ROUTER

In the research work, routers are configured for improving quality of service "best-effort"
packet forwarding. It means that the router forwards all packets it receives to the best of its
capability. The router forwards a packet where it can perform the table lookup necessary to
determine the appropriate interface for the packet. If the router is unable to send a packet
immediately, the packet is queued. If the queue is full, the packet is dropped. Packets are
typically processed on a first-come, first-served (or FIFO, First In First Out) basis. This
activity is adds-up to best-effort forwarding.

18
 Everything is typically good with best-effort forwarding until an interface is
oversubscribed. Once that happens, even if the oversubscription is momentary, the router
must queue packets to avoid dropping them. Hence the amount of queuing available on an
interface determines the amount of momentary oversubscription that the router is able to
tolerate on that interface without dropping packets and causing performance degradation.

Oversubscription of 1Gbps or 10Gbps interfaces is typically momentary - the bulk of

the network traffic is composed of internal flows which consume a large amount of
bandwidth in a small number of flows, and once those flows encounter packet loss they
collapse and stop consuming bandwidth.

R&E networks are sized for the science flows, so the smaller flows do not typically saturate
interfaces. However, there is often enough background traffic that the bursts associated with
high-speed transfers can cause those transfers to collapse as the large transfers momentarily
oversubscribe an interface and overflow its output queue. Since TCP performs poorly in the
face of even a tiny amount of packet loss, it is very important to configure routers and
switches with sufficient output queuing to accommodate the momentary oversubscription of
interfaces that comes with the bursty traffic patterns inherent in wide-area, high-performance,
TCP-based data transfers.

Consider 10G DTN node that is connected to a switch that has a 100G uplink. Due to the
speed step down, buffers are going to be required on the output queue of the 10G interface
that faces the DTN. If a higher capacity DTN upstream begins sending data to our 10G DTN
(e.g. greater than 10Gbps flows), this switch will absorb the additional traffic in buffers, and
if none are available will begin to drop the data adversely impacting TCP and network
throughput.

Some vendors will tell you not to increase the output queue depth on an interface, because
they base all their assumptions on a traffic profile that is very different from an R&E traffic
profile. They will tell you that it's not good to increase the output queue depth, because traffic
will fill that queue and this will result in increased RTT for traffic that traverses the interface.
This may be true for steady-state traffic (e.g. 10G of videoconferencing or VOIP, or millions
of web browsers), but this is typically not true for R&E traffic, especially the large data flows
associa,uted with the transfer of large-scale science data sets (e.g. supercomputer simulation
data, high energy physics data, high-resolution telescope data, etc).

19
High-throughput TCP flows as seen in an R&E network typically burst at wire speed even
though the flow will not run at wire speed in steady state over the duration of the transfer. So,
the network needs to be able to accommodate high-speed bursts during connection start-up,
but the link (and certainly the output queue) will not remain full because steady-state load is
not a full 10G. The goal of increasing the output queue depth is to give the network enough
"elasticity" to allow TCP to ramp up and smooth out without encountering packet loss early
in the transfer.

Optimum Buffer Size

Network gear with more buffer space typically is more expensive. Buffer is the basic
organization in the router. The general pipe line-rate of output queue buffer 10G switch, there
should be around 60MB of buffer. This is very important that to have a 10G host sending to a
1G host across the WAN. Here there are a number of switch design issues that make it hard
to quantify exactly how much buffering is actually required.

Figure 9

Network community and in the research work there has been a lot of discussion about Buffer
size and its working speed while too much buffering can be a big problem on networks with
speeds less than 100 Mbps. Problem at 10Gbps or higher is usually not enough buffering.

1.4.3Types of Routing Protocol

Distance vector Routing

It is used to find the shortest path between source and destination using lookup tables. It
transfers its updates to nearby nodes in every milli seconds.

20
Link state routing

Route of each nodes are updated in each router. Dijkstra’s shortest path algorithm is used to
find routing address in wireless nodes

ROUTING

DISTANCE LINK STATE

VECTOR ROUTING
PROTOCOL PROTOCOL

Figure 10

Enhanced Interior Gateway Routing Protocol

It is multicast routing protocol and well performed protocol which is considering the gateway
designing worldwide connections.

Energy Efficient Multicast Routing Protocol

The performance and energy efficiency of EE-MRP routing protocol is normally very good
protocol in all aspects. On the basis of simulation results, it is observed that routing protocol
(EE-MRP) has performed well in terms of overall network lifetime, throughput, and energy
efficiency. So in this research router Cisco 2010 Connected Grid Router, EE-MRP Routing
protocol is used.

1.5 Quality of Service Concepts

QoS enables us to provide better service to certain flows in the network. It can be achieved
by raising the priority of a flow or limiting the priority of another flow and to avoid the
packet loss or packet discarding.. While using congestion-management tools, the priority of a
flow by queuing and servicing queues in different ways. The queue management tool used for
congestion avoidance means preventing the congestion raises priority by dropping lower-

21
priority flows before higher-priority flows. Policing and reshaping is the queue provide
priority to a flow by limiting the throughput of other flows. Link efficiency tools limit the
large no of flows to show a preference for small flows.

1.5.1 QOS Architecture

End to End between network elements by identification and marking techniques.

 Single network element - queuing, scheduling, and traffic-shaping tools

 Policing , management, and accounting functions to control and administer end-to-end
traffic across a network

QOS Identification and marking

End to End network elements service to provide type of traffic is identified. Queuing is
marking by priority queuing (PQ) and custom queuing (CQ).Common methods of
identifying flows include access control lists (ACLs), policy-based routing, committed access
rate (CAR), and network-based application recognition (NBAR).

Congestion Management

An account of the bursty nature of voice/video/data traffic or real time application during this
time the amount of traffic exceeds the speed of a link. At this point Congestion-management
tools address these questions. This tool includes priority queuing (PQ), custom queuing (CQ),
weighted fair queuing (WFQ), and class-based weighted fair queuing (CBWFQ).

Queue Management

Queues are not of infinite size, they can fill and overflow. When a queue is full, any
additional packets cannot get into the queue and the additional packet will be dropped called
as tail drop.

Here not only dropping packet because of queue filling, apart from that when the packets are
standing on the queue buffer unit are getting thermal dissipation and it discard 10 % of
packets. Router cannot prevent this packet from being dropped tail drops .As to prevent
high-priority packet. So, a mechanism is necessary to do two things:

1. The queue does not fill up, but it is high-prioritized packets.

2. Allow some sort of Thermal management criteria to maintain for Packet dropping.

22
1.5.2 Throughput & Link Efficiency

Packet size: 8500-byte packet * 8 bits/byte = 12,000 bits

Line rate: 56,000 bps

Result: 12,000 bits/56,000bps = .214 sec or 214 m sec

Consideration low-speed links present an issue for smaller packets. The serialization delay of
a 1500-byte packet on a 56-kbps link is 214 milliseconds. If a voice packet were to get behind
this big packet, the delay budget for voice would be exceeded even before the packet left the
router!

Traffic Shaping and Policing

Shaping is used to create a traffic flow which limits the full bandwidth potential of the
flow(s). To prevent the overflow problem in the buffer unit mentioned. The central site
normally has a high-bandwidth link, while remote sites have a low-bandwidth link in
comparison. Policing and traffic shaping it is possible for traffic from the central site to
overflow the low bandwidth link at the other end. The perfect way to pace traffic closer to
384 Kbps to avoid the overflow of the remote link. Traffic above the configured rate is
buffered for transmission later to maintain the rate configured.

Service levels refer to the actual end-to-end QoS capabilities, meaning the capability of a
network to deliver service needed by specific network traffic from end to end or edge to edge.
The services differ in their level of QoS strictness, which describes how tightly the service
can be bound by specific bandwidth, delay, jitter, and loss characteristics. Network
communication, including lack of Bandwidth, delay, jitter or even loss of data during
communication.

Bandwidth

It refers to the speed of and capacity of a link, measured in bits per second (bps). It means
how many bits can be sent over the link per given second.
The networking devices QoS tools determine and control what packet is sent over the link at
a given source and destination; which messages get access to the bandwidth next, and how
much of that bandwidth (capacity) each type of traffic gets over time.

23
The WAN edge router link might be configured with a QoS queuing tool to reserve 50
percent of the bandwidth for very important or emergency data traffic, 10 percent for voice,
and leave the rest of the bandwidth for all other types of traffic.

1.5.3 Jitter & Delay

There is two type of delay here; one-way delay or round-trip delay.

One-way delay describes the time lapse it takes for a packet to arrive at the destination host.

Round-trip delay measures the time lapse between when the packet gets to the destination
host and the receiver to send it back.

So many different individual actions cause the delay of packets on a link; so many factors
cause delay when the packets driving from point X to Y...…

Consider the packet movements X,Y, Z in the network.

Packets in the network

Figure11

Packet X to Y is 210 ms delay. Packet Y to Z 225 ms delay. Packet Y to next node 230 ms
delay. so, the packet movements are changed in the network15ms ,5 ms delay or time
variations in the packet movement in the network is called as Jitter. This research works
under high QoS logic; hence jitter can be avoided and focused on packet drop due to varies
criteria.

1.5.4 Internal Section Of Router Unit

Router is the basic device broadly present in switching unit of the network. It is broadly
participating vital role to forward the packets to the right destination by the shortest path.
Whenever the packets are forwarded, it has been undergoing different internal functions
before forwarding the packets to the right location. As the above explanation mentioned the

24
packets are discarded due to several reasons however this research work mainly focused
packet discarding scenario due to router imbalance strategy due to power consumption and
buffer capability which unable to handle large no of packets cause traffic or congestion in the
network.

Flow Application Send to

Fragment
Receive Protocol & host
reassembly matching
Packet Rule
of IP table Matching
Packet

Send to Receive
host Packet

Figure 12

The above figure shows the internal configuration of router unit and its buffer functions. Each
unit task has been keenly observed the following i.e. a)Internal Packet Movements b) Packet
discarding ratio in each unit c) Reason of packet discarding in specific unit. This can be
elaborated later.

1.6 THERMAL MANAGEMENT

1.6.1 Silicon

It is well known chemical element is Silicon which has the symbol Si. It has the atomic
number 14 and it is a tetravalent metalloid and semiconductor. Is a chemical element with
symbol Si and atomic number 14. As per the comparative statement from several genius for
germanium and silicon, It has been identified that Silicon is more widely used semiconductor
for electronic devices, because it can be used at much higher temperatures than germanium..
Its melting and boiling points of 1414 °C and 3265 °C respectively.

25
 Atomic number -14

Atomic Weight-28.086

Si
Densitv(g/cm3)-2.33

Melting Point(C)-1414

Figure 13

As the silicon is having above fixed character, apart from that so many researches has been
done for how silicon is used to control thermal dissipation in electronics devices. It is
internally designed to provide heat transfer path between heat generating components and
heat sinks, heat spreaders and other cooling devices.

SILICON THERMAL MANAGEMENT BENEFITS

 Good thermal conductivity (1.6 W/m-K)

 Good converting for complicated shape
 Hardness: Shore 00 = 58
 Good flexibility with over bending
 Good dielectric properties
 Thin thickness, 0.20 mm and 0.25 mm
 High temperature resistance

Thermally Conductive Silicone Property Value Method

 Thickness (mm) 0.20 / 0.25 (±0.025mm)

 Thermal Conductivity (W/mK)* 1.06

 Flammability UL 94 V-1

 Density (g/cm3, @ 25°C)* 2.4 (±0.25)

26
  Hardness (Shore 00)* 55 (±10)

 Volume Resistivity (Ω-cm)* 8.6 x 1013

 Dielectric Strength (kV/mm)* 14.15

 Dielectric Constant* 100 Hz 1 Khz 1 Mhz

Hence this element can be well used to control thermal dissipation in all kind of electronic
devices.So far it can be used widely to control overall heat dissipation controlling
element.However in this research work silicon sheets can be used internally in the buffer unit
of router to managing packet flow when heat dissipation is the only reason of packet
discarding in the router.

Thermal balance Silicon Sheets

Silicone-based sheets from WACKER are used as thermal balance meterial in this research.
Thermal energy application range from -50 to +150 °C.When its dielectric element range is
increased propotionally to the heat dissipation factor.
Di ᵡ Dp
environmental conditions..This thermally conductive silicones Sheete protection are also
applied to the heat sink as a paste.Under low pressure, they can be flowed to form a thin.

Main features of silicon sheet

Table 3

Dielectric Strength Volume Thermal Volatile Content Thickness

[kV/mm] Resistivity Expansion Weight Loss
IEC 60093 Coefficient 150 °C/1,000 h
[Ωcm] [μm/m] or [%]
[ppm]

27
> 23 > 1014 180 <1 .05mm

The thermally conductive silicon sheets to the Router unithe part of figure ().In this
connection,sheets are specially formulated as per the features given in the above tabulation
are gentelly fixed on the reveiw unit.
If possible the sheets are pressed on the printed circuit board into the heat-transfer material.
It has been tested under room temperature.

1.6.2 SILICON OIL

Application
Silicone oil has been used for the protection of carbon steel alloy samples in a study. Silicone
oil has also been used as a heat reservoir during NMR (Nuclear Magnetic Resonance)
analysis.
General description
Silicone oil is useful for the preparation of oil baths having a temperature in the range of -
50°C to +200°C.
Aspect transparent Dynamic Viscosity at 25°C 50 mPa.s Density at 25°C 0,96 g/cm3 Tensile
Strength 0,021 N/m Flash Point > 250 °C Refraction Point a 25° C Approx 1,403

1. Temperature Change: silicone oil reveal a much smaller degree of change over a wider
temperature range than non‐silicone fluids ( Low Viscosity)

2. Thermal Stability: Silicone oil shows excellent stability when exposed to high
temperatures for long periods of time.

3. Chemical Inertness: Silicone fluids are chemically inert to most common materials of
construction.

Figure 14

28
4. Low Flammability: The range of 205ºC to 260ºC (400ºF to 500ºF) and auto‐ignition
temperatures in the range of 438C to 460C (820ºF to 860ºF) are typical for silicone oils.

5. Non Corrosive: Silicone fluids contain no acid producing chemicals to cause staining or
corrosion.

6. Low Surface Tension: Silicone fluids have unusually low surface tensions, which help
provide easy and efficient spreading, high surface activity and low internal cohesive
energies. Lower viscosities provide lower surface tensions than higher viscosities.

7. Dielectric Properties: Electrical grade silicone fluids offer excellent dielectric properties,
which are maintained for prolonged periods, even under adverse operating conditions

8. Specific Heat: Specific heat of most silicone fluids falls in the 0.36 to 0.39 BTU/lb./F
range at room temperature. The above silicon oil is convenently used for controling the
thermal dissipation.

1.7 MOTIVATION OF THE RESEARCH

Current scenario in world wide review electronic devices are normally undergoing the heat
dissipation problem while transmitting heavy no of mutimedia message in the
tnetwork.Unfourtunately TCP is impacted by this heat dissipation,(i.e) 34 % of packets are
affected in the means of thermal energy problem.Hence whenever the packets are
retransmitted due to Packet discarded senario there are heavy amount of impact on TCP
congestion window size which will not permit an optimal on throughput ratio.And also it
delivers the perception of limited capacity on the route by the sender. When latency is high in
the network, that the sender spends idle time (Some sender not sending any new packets),
which reduces how fast throughput grows.

Table 4

Round trip
latency under TCP throughput in the TCP throughput with
packet network 2% packet loss
deliverance

0ms 93.5 Mbps 3.72 Mbps

30ms 16.2 Mbps 1.63 Mbps

29
 60ms 8.07 Mbps 1.33 Mbps

90ms 5.32 Mbps 0.85 Mbps

Above table shows the irrespective of the reason there may be the losing packet in message,
acknowledgement, synchronizing.[65]This may cause due to Genuine congestion, Server and
electronic devices heat dissipation issue, packet shaping and aligning.[issue This will apply
irrespective of the reason for losing acknowledgment packets genuine congestion, server
issue, packet shaping. In order to avoid this latency due to impact on electronics devices is
the goal of the research.

1.8 Research challenges

An electronic systems one can act as specialist system must be capable of balancing to
workload variation, environmental changes, or even changes in overall hardware systems,
such as computers, Embedded systems, Artificial intelligent systems. As in computer
communications there are many challenges have been facing to get high performance
network. It has been focused hardware design, Jitter avoidance, several algorithms for routing
protocols, bandwidth allocation, Integrated and differentiated services, delay performance,

30
etc. In the hardware design the intelligence focuses to minimize the hardware element rather
for the future trends. Managing thermal management in hardware device to enhance QoS is
the current focus in data centers. Whenever the infrastructure is focussed such as
guarantied elements, device adopt for variations in environment changes and scaling issues, it
is critical to design. However, this is a challenging task undertaking due to the inherent TCP
packets path of design issues that arise from Physical level and Logical Level.
And also three-dimensional integration is a emerging technology that offers many
advantages. However the functions of heat energy needs several higher things to develop the
functions it is too expensive (69).
Now this research challenges are fully related to packet drop variation in multiple nodes can
be used for wired and wireless networks.

1.10 OVERVIEW OF THE PROPOSED ENHANCED VIRTUAL ROUTER

Current research work focus on to improvise upstream and downstream bits in the
range of Tbps, without any loss of packets in the network. This methodology is internal
arrangement of hardware devices in router/buffer in order to enhance communication
network functionality.

Figure 15 Figure 16
In the above figure 15 & 16 Modelling buffer size, there are seven units present in the
function. Define Subset data is the first logic when data enter the router. Data are defined by
IP addresses and are forwarded to the flow match configuration. Here same IP address
packets are segregated in call the batch match logic data. In this unit group packet are
analyzed and aligned. Aligned packets are executed and mismatching IP Addresses packets
are dropped out. During this time bottle neck occurs in buffer unit, yellow coloured defines
31
the dissipation of heat energy in figure (2). In order to enhance the performance of the Buffer
unit, Special design called path diversity are used in Review batch section shown in figure
(2). For the eastward source-destination pairs, such as S3 − D3 andS6 − D6, the east-north
(EN) turn and east-south (ES) only allowed at even columns. For the westward pairs, such
as S4 − D4 and S5 − D5, the west-north (WN) turn and west-south (WS) turn are only
allowed at odd columns. Path diversity is the basic mechanism used in review batch to reduce
heat dissipation. The power profile medium thermal limits, setting suitable D is essential to
improve the thermal-limited throughput. Because the vertical traffic loading is redistributed
by the proposed routing-based traffic migration method, the channel loading will be
unbalanced. we will propose the vertical buffer allocation method to solve this problem, and
we introduce our systematic design flow to find the most suitable downward level With
increasing thermal energy are proportional to throughput and the performance of the TCP
increases.

2 REVIEW OF LITERATURE
2.1 Introduction
Packet drop is the common scenario faced in electronic devices of data
communication multimedia messages. This section explains about the functions of avoiding
packet drop in the network due to thermal leakage. Meaning of avoiding thermal leakage
causes the packet drop in the router referring the method silicon based balancing unit, silicon
oil sheets based balancing unit, path diversity applications in the network, Which available
wire based network (WBN )and wireless based network(WLBN).Earlier detection can be
slight changes in the dropping function due thermal dissipation using various protocols,
enhanced design equipments, proper frequency and bandwidth maintenance, monitor based
congestion and collision. Now this following study helped to develops the designing part of
enhanced virtual router (EvR) which is deployed by keenly observing of previous references.
Study and references focused on silicone based thermal reduction on electronic devices.
Evaluation Parameters are observed and comparative studies made for each thermal
reduction router unit.
The temperature of the current router and the temperature of the memory and the
processor in the current tile. The work proposed the thermal coupling effect of the lateral
neighbour memory and processor of the vertical aligned routers[1]. Sundar Iyer proposed
(presumably on-chip) SRAM cache that grows in size linearly with line rate and the number

32
of queues, and decreases with an increase in the pipeline delay, hence heat dissipation is
constantly maintained[2]. The necessity of silicon oil in the thermo dynamics element and
how temperature is maintained in dielectric material using silicone oil sheets [3].Jennifer
Casella proposed an alternative to the traditional method of IP routing known as flow-
based routing, whereas traffic is sent across the network as part of a common flow, rather
than individually inspecting each packet [4]. Sudhir N. Shelke proposed to avoid jitter
variation among the packets in network pipeline concepts were introduced in the internal
circuitry of router which avoids flawless communication. Enhanced virtual node stream is
adopted by this author[5]. Karlapudi Sravani, P T Balakrishna proposed to system on chip
with multipurpose networking router to enhance the packet drop technique and it is tested by
buffer unit test[6]. Chrysostomos A. Nicopoulos that to his unified buffer structure called the
dynamic Virtual Channel Regulator (ViChaR) which dynamically allocates buffer resources
according to network conditions and reduced packet standing in the queue. [7]
However LPWANS covers the packet deliverance in the Limited spanning area for
over 520 meters. When its spreading factors 868MHZ ISM band, its packet deliverance
ratio was 96.7%.[12]. ATPG Technology is systematic debugging packets of each unit.
Forwarding table verification unit specifies and generates high heat dissipation 94 degree c
[21].
The detection packet method is analysis by applying statistical path detection. Each data
communications devices packets are traced in the means of compressed. It is for tracking
packet loss comprised in the single statistical unit which means for selectively identifying a
router interconnection of router interconnections for packet irregularly. Count can be
detected. The tracking packet periodically exchanging statistics development of packet
anomaly count detection to synchronize the count in each unit specially in counts between the
termination points of the router interconnection and a particular opposed termination point.
Here Sequence number in a packet has been highly focused the sequence number
corresponding to a set of packets. The packets are identified by the means of posting statistics
to a remote PE router by at least one of a VCCV status, L2TP WAN error notification, BFD
keep alive, and an LSP status message.[13,54,59]
Hardware element packet transferring is over-the-air a Radio Link Control (RLC) and the
status report from a user equipment (UE) to a target node, Here RLC initializing the messages
what to be received at the source end. If any packet is dropped RLC PDU method, sent the
dropped sequence nearer node has introducing protocol segmented Packet Data Convergence
Protocol (PDCP) Service Data Units (SDUs).Acknowledgement packet & requesting packets

33
are sequenced where used the same protocol. It is best functions of identifying the lost
packets and to forward again.[14,30.38]
RE-ATTEMPT routing protocol for Wireless body Area Sensor Network (WBASNs).Here
single-hop and multihop routing are used. Specific analysis is based on Priority based route
selection for the delivery of normal and emergency data. The first point of study is based on
multihop routing is delay in which the data were standing in the queue for the long
time(20ns); Second report is based on energy consumption based packet drop in multihop
model. RE-ATTEMPT protocol in used as path loss of finding problem formulation and its
solution via linear programming based mathematical operation. It is also analysing energy
distribution of the node for the each node with the multihop equation 𝐸𝑀-HOP = 𝑛 × 𝑏 ×
(𝐸𝑡𝑥) + (𝑛−1) × 𝑏 × (𝐸𝑟𝑥 + 𝐸𝑑𝑎)[15,19,58]
Energy consumption of router frequency and power scaling management system. It refers the
policy of thermal dissipation cause of high power consumption in the router. Author
focussed high power deliverance and its main causes of the router. It is introduced level
intrusion detection systems (NIDS) which are keenly focussing packet discarding scenario
due to high power functioning. In order to reach less discarding packets a low-latency and
energy-efficient NIDS (LEoNIDS) is effectively used.[16,25,35]

Thermal rise in electronic devices affects the data communication devices one of them are
called routers. It is proving that the beneficial necessary of Quality-of-Service (QoS) in
the router. Thermal rise in routers are identified and it’s no of packet discarding is reduced
by using multiple Software. TLQOS Software enhancement provisioning is to reducing
packet drop issue for an traffic types, as well as avoids the formation of highly heated nodes
known as hotspot(s), and keeps the temperature rise along the network to an acceptable level.
Simulation result shows TLQoS exploits the performance of modular protocol design used to
proves the small no of packets discarding. [17,21,22]

The functions of retransmission of discarding packets due to thermal energy in the router. If
the packets are discarded due to this flaw Selective Discard & Retransmission of packets
mechanism are used find outs the packet sequence number and its better to retransmit the
data. SD & RP mechanism divides the packets according to the sequence number and finding
congestion free in the network Proposed design is enhanced the high efficiency of throughput
in the network.[20]

34
Semiconductors have resulted in substantial increase in energy consumption, subsequently,
heat dissipated from data center racks. It has been proposed to overcome these energy
efficiency and thermal management challenges. The performance of the data centres highly
choosing the combination of energy reduction. The energy and cooling efficiency of a small
data center is then evaluated by applying several metrics. [31]

Adaptive routing algorithm software management of packet dropping is wisely chosen to

reduce. Traffic- and Throttling-Awareness Routing (TTAR), to tackle the traffic congestion
due to throttling of transient-temperature control. The experimental results that TTAR not
influence the capacity of heat dissipation seriously, Meanwhile it can attain 8% ~ 680%
throughput improvements than the preceding routing algorithms at 50-cycle average
latency.[33]

Three ILP-based static thermal-aware mapping algorithms for 3D Networks on Chip (NoC)
to explore the thermal constraints. Two approximation algorithms and analyze algorithmic
complexities to shows its proficiency. Mapping algorithm employs to minimize temperature
up to 24% and 22%, Mapping algorithms does not affect the chip temperature. [34]

Networks on chip (NoC) and 3D integrated circuits have been proposed as solutions to avoid
thermal dissipation on the electronic devices due to the increase in the power density of parts
of the integrated chip. Genetic algorithm for a thermal and communication aware mapping
are the basic algorithms used for reduction in system temperature when compared to a
random mapping.[35]

Introduced the proposed energy control systems design which provides intelligent services
for users and demonstrate its implementation using a real testbad.[36]

Proposed myriad techniques to curb the energy consumed by nodes. Here, it emphasizes
distributed system level and trying to improve aspects such as resource allocation,
scheduling, and network traffic management in order to enhance quality of services (QoS)
[53].Apparatus including a package configured to mate with a printed circuit board which is
explicit control method of thermal management [47].

Focussed on multiprocessor internal router functions used in Network on Chip (NoC)

infrastructure. Quality of service experienced of the router to avoid congestion and to ensure
efficient data-flow control. Maintaining an interesting hardware characteristics which
designed with 45 nm integration technology.[48]

35
 An Analysis of flow-based routing .This paper deals with the router unit 0.000022 seconds,
as compared with about 0.000018 seconds in traditional IP routing. Hence packet flow is
monitored and drop were calculated by the method of flow based routing. Traffic 0.51
seconds[4,59]. The first part of the review includes an introduction to different approaches
the need of thermal management in the modelling data centres. Experimental studies are
reviewed in the second part, followed by a review of recent technologies in enhancing
thermal management of data centre.

2.2 Summary

Table 5

Author(s) Year Thermal management Methods Packet

of router discarding
Percentage
John G. Apostolopoulos et al 2001 Path diversity 50%
[19]
Luca Benini [38] 2002 System on Chip (SOC- Not Specified
deterministic and stochastic
models)
S. Kalyanaraman [46] 2003 heuristic search algorithm, 33%
Recursive Random Search(RRS)
B. Quoitin et al [45] 2004 adaptive segmentation of Packet 44.5 %
traces

Addo-Quaye et al[35] 2005 genetic algorithm on 3D NoC Not specified

environment.
Chunsheng Liu et al [96] 2006 dynamic thermal-balance routing Thermal
(DTBR) algorithm dissipation220v
Amy R.Reibman et al [63] 2007 SSIM, MSE, and a slice- Not specified
boundary mismatch (SBM)
metric
Ayse Kivilvim Coskun et al 2008 multiprocessor system-on-chips Normal thermal
[51] (MPSoCs) dissipation160 V
Packet drop 52%
Yu-Wei Yang et al [54] 2009 efficient dynamic frequency-to- Not specified
voltage scaling (DF2VS) algorithm

36
Shervin Sharifi et al [56] 2010 TEMPB -Thermal and energy Thermal
management technique Power balanced
balance
Martin Suchara et.al, [21] 2011 Overall internet routing network 2% improved
comparison studies. Rather than
software implementation,
Hardware thermal Management
can be used.
John H. Lau et al [66] 2012 Junction temperature and thermal Reduces .859
resistance of 3D IC SiP with heat dissipation
various TSV interposers
Konstantin Avrachenkov [23] 2013 Drop Tail buffer management Thermal
policy Buffer size can be set by balanced-
applying round robin method of 66Mbps through
function Put of 50v

Anne-Cecile et al [54] 2014 Network virtualization Not specified

Noris Gallandat et al [32] 2015 Distribution scale power routers Reducing 3 to 5
(13–37 kV, 1–10 MW) % of energy

3. FLOW BASED ROUTER

An examination of techniques for avoiding thermal energy leakage in sensor networks is the
key design system factor in communication networks. It has become a major concern in
other systems. Research and development had been mostly driven by performance of
networks. Gain in performance of network by minimizing packet discard, whereas rises in
improving throughput. Deploying upgraded virtual router design enhances the hardware
technology of energy consumption and dissipation by the means of heat energy. In order to
improvise router performance, this research focuses the hardware structure of router power
consumption and energy dissipation. Hence, energy efficiency of the nodes is the common
field to find out computing resources, storing data and its movements, forwarding technique
in the router.
.
3.1 Energy Efficiency of internal nodes
Energy Efficiency of internal nodes

Measuring power Node level Grid and data centre Cloud and Virtualization
Consumption Peak points power management

37
Hardware Capabilities Sleep states Dynamic voltage Software improvements
and frequency scaling

Electricity efficient sources Thermal management Workload consolidation Task

scheduling
Fig : 17 Overview of techniques to improve the efficiency of internal nodes
In the above figure 17 it is mentioned that the major techniques to emphasise the
communicating internal nodes efficiency by the means of different area. Measurement of
power consumption in the hardware components of computing resources can be resolute by
energy sensors example: wattmeter’s or be estimated by energy models. Power Distribution
Units (PDUs) is embedded and it can be measured by temperature sensors. Microsoft's
Joule meter can also used for software tool that evaluate the power consumption
Power Usage Effectiveness (PUE)
PUE = Total Facility Power / IT Equipment Power

Node level Peak Points are called as node optimization which normally comprises with
several components, each one designed to save energy. That can be utilized in two ways 1)
Static 2) Dynamic. Static can be the node either used in forwarding techniques or used in
different sleep state. If the node is nearer to the destination that node can be used for
communication called as compute working nodes (CWN). In the sleep node the node can be
the energy free node. Dynamic nodes are dynamically fixing the frequency according to the
incoming packets. Normally these nodes are used for all stages, but its frequency & energy
consumption varies according to incoming packets. This research focuses dynamic node
thermal management.
In grid and data centre focuses power management that is saving energy at the nodes. When
these nodes are keeping nearby any electrical source transmission loss can be reduced. Hence
power management and thermal management are very important to maintain the energy

38
efficiency of computing resource in the data communication (DC).It can be called as green
sources of electricity.
In thermal management, thermal dissipation and its related issues are the most direct
consequences of infra structure of the design of processor chips. These issues are interrelated
as decreasing heat production reduces energy consumption. To enhance power efficiency in
the nodes many algorithms deal with both energy and thermal problems .An infrastructure of
PUE, that cooling itself consumes quarter the amount of power by the computing resources.
Thermal load balancing can be used to reduce cooling devices power consumption.
Workload consolidation is the techniques where it consists of multiple tasks in the same
router. (i.e can be used for switching, bridges, links, etc).

Below table shows the variations in power cooling devices cost of mentioned years.
Whenever the technology upgraded in the DC, computing nodes working under heavy load
which causes to enhance the design of cooling material. Such a way that cooling material cost
is also increasing.
Table 6

year 2004 2006 2008 2010 2012 2014

Power and 15 20 25 29 38 52
cooling cost in
percentage

Power and cooling cost in percentage

80
60
52
40 38 Power and cooling cost in
25 29 percentage
20 20
15
0
2002 2004 2006 2008 2010 2012 2014 2016

Figure 17
3.2 Reason and modelling of energy leakage

39
In the above figure the conclusion of studies says about the growth of computer network
technology may cause the variation in the thermal energy of the devices. Whenever this up
gradation of technology infrastructure improvises the efficiency of power management in
the node, the communication will achieve full-fledged 100% error free approach. Internet
accesses thousands of public and private nodes and linked by various hardware devices
like electronic components and optical devices. Messages are passes through such
heterogeneity environment and its function constraints make it difficult to forward the
message in right location. Hence it is precise to estimate energy consumption and leakage
of each node in the network is very important. At the same case, measurement of energy
consumption in each unit of router and its performance details carried out in this research.

Energy
consumption
efficient
Thermal Router Packets are
leakage standing in
discards the the Queue
packets stand in
the queue

Raised Router Pressure

pressure is raised due to
delivers heavy queue in
thermal Tbps of
leakage transmission

Figure 18 : Steps in reason of packet discard

In the above figure there are 5 steps which show the heterogeneous network and dynamic
router performance. An efficient router has the internal steps, while processing in forwarding
the packets in Multicast algorithm. Multicast algorithm is common for all type of nodes.
TCP/IP protocol is the internet protocol where used to govern the traffic and scheduling
policies. TCP protocol always tries to fiddle with the sending rate of a source to match the
available bandwidth along the path. If the congestion signals are not forwarded the network
TCP/IP congestion window increases gradually and TCP reduces the congestion window the
reception of a congestion signal,
 Step 1

40
 Router is the efficient device which is designed to maintain the flow between end to
end points of heterogeneous networks. It carriers out the mechanism of Static and
dynamic router and in round trip method (RTM). It is purposefully designed to avoid
anomaly of the packet movements.
 Step 2
Incoming packets are forwarding through buffer unit which designed internally in the
router node.TCP/IP creates backward Explicit congestion control (BECN) and
Forward explicit congestion control (FECN) notification in the buffer unit when the
packets are dropped. During the end to end Tbps of data transmission, there may be
millions of packets are moving in real time applications. Hence, in the buffer unit
packets are standing in the queue to verify and forward the packets in right IP address
nearer node to reach the destination. These large no of packets in the queue raises
electronic devices usage. For the long time pressure in electronics components,
increases heat dissipation in outer part of router.

Figure 19 : Cross view of Router cooling

It is effortlessly adjustable to deliver the best wireless signal .Deployed advanced heat
dissipation with high- density aluminium coated and optimized performance and heavy
load stability.
 Step 3
Buffer unit packets are standing in the queue due to forwarding verification process.
When the packets are moving multiple path it is undergo bottleneck problem. Such issues
may cause heavy pressure in Tera bits of transmission. Now the nodes undergo high
pressure and delivers heat energy.
Suppose there
K flows trying to deliver packets to their destinations via a bottleneck router with the

41
 Following parameters:
C is the bandwidth, (in packets per second);
The flow starts in the deterministic initial congestion window value nk and it always has
packets willing to deliver. For model transparency we ignore timeouts (and the corresponding
slow-start phase of TCP), since they are known to be significantly less frequent than
congestion events and we expect they would not affect our solution significantly.
 Step 4
Raise in pressure creates heat energy dissipation. Now the node buffer unit is full of
heat
Dissipation pushes the device to unstable condition though it has good efficiency.
 Step 5
Due to this unstable condition, the packets are becomes weak in frequency range after a
nanoseconds it will discard. Discarding packets sequence TCP/IP windows sending
congestion notification signal to the sender and receiver end. Network upstream and
downstream heavily affected by collision and congestion is the main reason of delay
transmission.

3.3 Flow based Buffer flexibility in IP

Flow based packets are permit both flow-based and packet-based forwarding
simultaneously on a router. In order to proceed and avoid traffic in the buffer unit special
function of packet-based, forwarding is used in stateless firewall filters, which are known
as access control lists (ACLs). Whenever the look up tables are monitored, by passing flow
based forwarding and introducing IP based forwarding is the best performance of the router.

By default, formation of the node, packets entering the system are the device for selective
stateless packet-based processing; packets entering the system are assumed differently:

 Whenever the flow packet based functions are applicable for the incoming packets it is
marked as IP forwarding algorithm. 4Gbps router design satisfies the process of policing and
classification according to next nearer router.
 Buffer utilization of this part is the common scenario of the normal processing. If a packet
has not been marked packet is consider as flow based and it develops priority less queue.

42
 Whenever the packets are under perception mode, it undergoes normal processing. All
services can applicable for except for MPLS can be applied to this traffic.
 Traffic flow with selective stateless packet-based services bypassing flow-based processing.
 By passing look up table and categorizing the packet with marking IP address and queue is
scheduling & policing.

Forwarding look

1 2 3 4 1 2 3 4
ROUTER
Figure 20

But most of the time forwarding table bypassing flow is critical to locate the packet in the
correct destination.
When the packets are entering inside the buffer without by passing the forwarding table
buffer unit follows active queue management (AQM) which controls the amount of data
stored in network node buffers. In some unavoidable situation AQM is technique where
preventively dropping packets depending local congestion. It can be used to mark the packet
instead of dropping immediately. After all evaluation a marked packet is silently discarded. A
set of services are needed to avoid this kind of discarding.a) IPv4 routing/TCP b)Link
fragmentation and interleaving(LFI) c)Multiprotocol label switching(MPLS) d)Compresses
real time transport protocol (CRTP)

3.3 Markov Decision congestion Model

It is the major mathematical formulation of controlling congestion problem using a Markov
Decision Process (MDP).It is allocating framework of resource allocation, which extends
the restless bandit model. Let us consider timeslot of the nodes with suitable functions, and
in case the drop packet creates the FECN and BECN formulates heavy congestion in the
network. Choke packets are moving both the end in the limited time slots, hence the 100
nodes affected by heavy congestion, in order to avoid that Markov calculation were included.

Congestion occurs in Router 2

43
 BECN FECN

Ack Ack Ack Router 3 Ack

A Router 1 Router 2 B

Packet 1,2,3.4 Packet 1,2,3.4 Packet 1,2,3.4 Packet 1,2,3.4

Figure 21
In order to find tractable method to identify congestion area a simple, implementable
solution will be used. It is considered as several simplifications of the features of the
problem. The performance of the solution is evaluated.

Let us consider the time slotted into discrete time 2 T := f 0; 1; 2; : : : g

Time periods of one round-trip time (RTT)
It is assumed equal for all the flows. However, in Section 3.2 this constrain disappears
and obtain nice results too. So, consider users with equal RTT, although it can conclude from
the simulations this constrain can be omitted. It is assumed that all packets are of the same
size, which it is further define to be one bandwidth capacity unit. The router takes decisions
about admitting or rejecting the flows at every time.
3.4 Packet arrival rate exceeds the outgoing link capacity
The sudden a large mass of data arrives from several input sources and this large no of
multimedia packets are needs to be forwarded through a common output link, then a long
queue of packets will be created at output MIME link. Congestion Control and its various
techniques are If the size of queue is large, the packets will be dropped.
If the router has infinite memory the situation may even get worse, for example if there is
along long queue of packets, the packets a tail end will take some time to reached the head of
queue to be processed but in the meantime, time out for majority packets occurs resulting in
useless processing and duplication of packets. Thus increasing the load needlessly and
making the situation worse. It was Nagel who proved that network with infinite buffer is also
susceptible to congestion as network with normal buffer. The size of buffer should not be too
large as it will create more delay and waste processing time. (a) Queue with infinitely large
buffer (b) Queue with small buffer Effect on packets of with large and small buffer.

44
Providing a high speed links seem to control the congestion but it is worse condition in
some situations. Since high speed balanced links and network processing need to be
configured. In reality it’s almost impossible.
If in case the Timed out packets Buffer Overflow Packet Discarding Chapter 4: Congestion
Control and its various techniques 81 balanced links with same speed are configured with
processing, still bottleneck possess problem. Bottleneck causes the significant degradation of
performance due to limited amount of resources available. In spite of the fact that all the links
are of same speed 30 Mbps, but at the bottleneck link the data is thrice the capacity of link
causing congestion and plunging the quality of the entire network. Representing the
bottleneck in the balanced network is the major concern of DC. Bursty field and slow
processor is also the function of Network when TCP/ windows become high collision and
congestion. Such kind of heavily spoiled network is slow down its processing speed.
Slow processor: The slow processor of routers can also cause congestion. If the processor is
slow in performing various tasks like processing queues, processing routing table etc. The
large queue will be still build, although having links with excess bandwidth. Increasing the
bandwidth of line will not increase the speed of CPU and vice versa. Therefore there is only
the shift of bottleneck from one part of system to another. With the real cause as the
mismatch between the various parts of the system.
4. VIRTUAL ROUTER

4.1 Virtual Router

In wire and wireless configuration the collection of immobile and mobile nodes are
communicating with each other. During their communication Packets were moving nodes to
nodes. First, Packets are standing in buffer and then enter to the nodes for Processing. Buffer
size is also one of the criteria to improve the Quality of service; hence the size of the buffer is
manageable. 1) In the manageable buffer size less no of packets were stand in the queue, rest
of the emerging packets were discarded 2) If buffer circuitry operations are long time heat
dissipates and packet may discard here. Because of this two criteria, The amount of packet
discarding is increased; In the result, the upstream nodes always receive the negative
acknowledgement from the downstream nodes, it creates heavy traffic like congestion and
collision in the network. Our main goal is to avoid the discarding large no of packets due to
heat dissipation of the buffer. To implement this service model we propose a dynamic
Provisioning architecture for intra domain quality of service in the buffer. Here the heat
dissipation ratio is identified according to the buffer size and incoming no of packets are

45
channelized. In the proposal we have to prepare efficient buffer enhancement that takes
explicitly into account traffic statics to increase the user benefit and the network revenue
simultaneously. This paper deals with hardware Technique of buffer and this would be
enhancing the performance of the network. Enhanced Virtual Routers maximizes throughput
on demand. EnVR ability to provide high performance with half the buffer size of a generic
router is of paramount importance, since it can yield total area and power savings within a
minimum area. Router buffers are important instrumental in the overall operation of the
network. As the Buffer plays vital role to forward from node to node power consuming is
high for down Stream and up Stream leakage power. These power consumption dynamic
changes may cause as packet flow throughput decreases. In some cases it has been observed
that storing a packet in a buffer for long nano seconds consumes large energy than
transmitting the packet .When Routers Plays a role it dominates buffer Size. Concurrently
buffer design plays a sensitive role in architecting high performance and energy efficient in
internetwork connection.. This paper deals with hardware Technique of buffer architecture in
router. Router area and power budgets are increasingly being dominated by the inter network
connections.Concentration focused on multi-core designing. So as to minimize
communication power has become comparable to logic and memory power. Researchers have
proposed sophisticated router architectures with performance enhancements, fault-tolerant
mechanisms. Area-constrained methodologies, minimizing delay, power-efficient and
thermal-aware designs is also a important design to deliver less packet loss ratio.
4.2 Thermal Management in Routers
Thermal management of routers are simulating high throughput in data centres called
network. The model is developed by considering a thermal leakage in router and its
implementation in new way of reducing the leakage. Some of the previous research used ten-
server rack technology and it predict an order of a 102–103 reduction an accuracy of 90%
packet transferring. Hardware features deployment can be enhanced thermal load
management, which improves the router heat dissipation capacity. The implementation is
trained virtual balance in temperature and distributes energy efficiency according to the
router thermal environment. Upgraded virtual router is a technique of analyzing flow of
packets movements and reducing packet dropping across a range of components in a system.
It is used to reduce percentage of heat dissipation and inner overall characteristics.
Compared to algorithm, software it is a relatively simplified method less expensive which
can produce accurate predictions within a short computation time.

46
Computational virtual router provides detailed descriptions of transport of packets in each
data nodes. These models are considered computationally intensive and are often
oversimplified, compared to other development and also it provides high accuracy in
reducing packet loss. These work finds that the cooling best utilized when the servers are
powered uniformly and are clustered inside a router. The average error in temperature varies
in the router between 7% and 12% at different elevations.

Liquid cooling systems are also used in implement level models of internal router. Here it
indicates that the change in temperature in the router unit. It finds that the internal
resistance affects pressure, temperature rises and consequent flow rate.

Router hardware elements are shown in below figure. As discussed chapter router is basic
part of the unit to transfer the packet from end to end source. However filtering and
forwarding most important functions of router. Let us take the router device Cisco 1120
connected in grid unit router and it includes external and internal division hardware units.
External unit consist the following unit 1) Router chassis 2) internal components 3)
Connectors 4) Ports. And also the internal architecture shows consists1) Subset unit 2) Match
configuration 3) Review batch 4) Create duplicate Resolution pasted.
This external and internal platform is built to meet the high performance communication
infrastructure needs to satisfy quality of service (QoS). It enables increased power grid
efficiency and reliability, reduced energy consumption, and unwanted emissions. The router
is also enables distributed intelligence like to marking the forwarding and filtering the
packets. The internal & external functions of routers shows the working Integration of
Distributed Energy Resources (DER).It provides secure and reliable real time communication
between data communication network system and the devices like distribution grid,
protection relays, distributed solar forms.
Comparative study made on which algorithm based thermal management and hardware based
thermal management. The results shows that the slight change of the saturation point on the
Prediction of flow rate in packets is important when obtaining physics-based representations
of data. Internal & external unit delivers cumulative pressure (i.e) internal static pressure in
the 250–500 Pa range and external static pressure in the 100–200 Pa range and the
temperature in pressure changes across the entire node. Hence it is proved that optimized
technology. In Chapter 1 figure 3 shows thermal leakage of unit in DC. 34% of thermal
energy is wasting in switching & routing device. By using wattmeter it is estimated that 16%
of internal section, 10% of external section, 8% of switching unit dissipates heat energy.

47
 Router element thermal Dissipation
18%
16%
16%
14%
12%
10%
10% internal section
8%
8% external section
6% switching unit
4%
2%
0%
internal section external section switching unit

Figure 22: Percentage of heat Dissipation in Router node

Above graph shows overall heat dissipation of router unit.It is used to develop predictive
statistical models for implementation in control schemes. Below figure shows the router
Configuration of internal and external. In chapter 1.5.4 it can explain internal configuration
of router (i.e) 1) Fragmentation and reassembly of IP packet 2) Flow matching table 3)
Application Protocol & Rule Matching. In chapter 1.10 figure 15 & 16 explains inner core
design of internal router 1, 2, 3.
External part of Router explains the interfacing part of router to the network. Parts are
1)receive packets 2) send to host In figure output port and input port is called as external part
which terminates the physical link and performs data link function. It can also perform look
up table forwarding routing protocol information passed to routing processor. Switching
fabric, router processor is the unit where it connects input and output port. It is core unit of
router which dissipates major area of heat dissipation happening in the node. Following
research serves percentage of heat dissipation in each internal and external unit.
Energy efficiency of electronic devices across the router is individually measured by
watt metre, dissipation energy calculated by the means of packet. Packet loss in the buffer
unit reduction is the main focus of this research.

48
 Figure :

External Router Internal Router

Figure 22 overview of router internal & external architecture

4.3 Cooling system failure

When the environmental temperature varies or module exceeds the temperature threshold,
the fans can formulate to operate high at full speed. Temperature and humidity Maintain
appropriate temperature and humidity in the equipment room.

1. There is an issue on poor insulation, metal corrosion, electricity leakage, mechanical

property change of materials,
2. Bring problem in including loose captive screws and circuit failure.
3. Insulation materials aging can create high temperature acceleration which causes
decreasing in reliability and significantly lower lifespan of the router.
Table 6 Temperature requirements of the router
Router Temperature
Operating temperature 0°C to 45°C (32°F to 113°F)
Storage temperature –40°C to +70°C (–40°F to +158°F)

Table 7 Humidity desires of the router

Router Relative humidity

49
 Operating Humidity 5%RH to 95%RH
Storage Humidity 5%RH to 95%RH

Electronics cooling system like heat sink, fan etc like hardware system command shows
following information appears on the configuration terminal:
%Dec 7 11:49:16:436 2016 HPE DRVMSG/3/TEMP_WARNING:
Environment temperature is greater than warning upper limit in
slot 0, index is 1.
#Dec 7 11:49:20:444 2011 HPE DEVM/1/BOARD TEMPERATURE UPPER:
Trap 1.3.6.1.4.1.25506.8.35.12.1.16: chassis Index is 0, slot
Index 0.0
%Dec 7 11:49:20:444 2011 HPE DEVM/4/BOARD_TEMP_TOOHIGH: Board
temperature is too high
on Chassis 0 Slot 0, type is HSR6602-XG.

For a period of time after the router runs, the environmental temperature of the router will
decrease below the high temperature threshold, and the following information appears on the
configuration terminal:
%Dec 7 11:50:11:436 2011 HPE DRVMSG/3/TEMP_NOWARNING:
Environment temperature recovered from TEMP_WARNING in Slot 0,
index is 1.
%Dec 7 11:50:11:436 2011 HPE DRVMSG/3/TEMP_SYSNOWARNING:
System temperature recovered from TEMP_WARNING.
#Dec 7 11:50:13:244 2011 HPE DEVM/1/BOARD TEMPERATURE NORMAL:
Trap 1.3.6.1.4.1.25506.8.35.12.1.17: chassisIndex is 0,
slotIndex 0.0
%Dec 7 11:50:13:244 2011 HPE DEVM/5/BOARD_TEMP_NORMAL: Board
temperature changes to normal
on Chassis 0 Slot 0, type is HSR6602-XG.
CPU temperature of a module exceeds the critical temperature threshold; the ALM LED of
the router is red.
%Dec 7 11:17:01:436 2011 HPE DRVMSG/3/TEMP_ALARM:
CPU temperature is greater than alarm upper limit in slot 0,
index is 0.

50
The high temperature threshold and critical temperature threshold of each module
1. Though cooling system is available in the router, it is under overall progressive, router
reaches critical temperature threshold value.
2. Though working environment of the model router is well ventilated, router oscillates
threshold temperature.

4.4 External Copper Ethernet port Router

External unit of the router connects the external network to router unit. It is pathway of the
router that transmits the packets from the gateway of the router to inner core devices. It is
only managing the router and it does not provide any servicing like data forwarding
capabilities. The management Ethernet port is a 10Base-T/100Base-TX/1000Base-T RJ-45
ports. It upgrades software and manages the router through a network management server
without using any service interface of the router.

Table 8
External port Device specifications
Connector RJ-45

Interface type Automatic MDI/MDI-X

Frame format Ethernet_II

Interface speed and duplex mode 10 Mbps, half/full-duplex

100 Mbps, half/full-duplex
1000 Mbps, full-duplex

Standard compliant Asynchronous EIA/TIA-232

Baud rate 9600 bps (default) to 115200 bps

Noise level 55 dBA to 65 dBA

Heat dissipation 102 BTU

British Thermal Unit (BTU) is the unit have been used to find heat dissipation in the port
unit. The media dependent interface (MDI) standard is commonly used on the port of
network adapters. The media dependent interface crossover (MDI-X) standard is commonly

51
used on hubs or LAN switches. Combo interface is the copper port, that either the copper
port or the fiber port. To switch between the copper and fiber ports it is used the combo
enable or { copper | fiber } command in interface view. This research measures the heat
dissipation only copper interface port. Following is the command for interfacing the ports
[Sysname] display interface GigabitEthernet 0/0/2
GigabitEthernet0/0/2 current state: DOWN
Line protocol current state: DOWN
Description: GigabitEthernet0/0/2 Interface
The Maximum Transmit Unit is 1500
Internet Address is 51.1.1.1/24 Primary
IP Packet Frame Type: PKTFMT_ETHNT_2, Hardware Address: 000f-
e200-0005
IPv6 Packet Frame Type: PKTFMT_ETHNT_2, Hardware Address:
000f-e200-0005
Media type is optical fiber, loopback not set, promiscuous
mode not set
1000Mb/s, Full-duplex, link type is autonegotiation
Output flow-control is disabled, input flow-control is
disabled
Output queue : (Urgent queuing : Size/Length/Discards) 0/100/0
Output queue : (Protocol queuing : Size/Length/Discards)
0/500/0
Output queue : (FIFO queuing : Size/Length/Discards) 0/1024/0
Last clearing of counters: Never
Last 5 seconds input rate 0.00 bytes/sec, 0 bits/sec, 0.00
packets/sec
Last 5 seconds output rate 0.00 bytes/sec, 0 bits/sec, 0.00
packets/sec
Input: 0 packets, 0 bytes, 0 no buffers
0 broadcasts, 0 multicasts, 0 pauses
0 errors, 0 runts, 0 giants
0 crc, 0 align errors, 0 overruns
0 dribbles, 0 drops
Output:0 packets, 0 bytes
0 broadcasts, 0 multicasts, 0 pauses
2 errors, 0 underruns, 2.5 collisions
0 deferred, 0 lost carriers.
Thus the above simulation represents the functions of external port and its heat dissipation.
5 Upgraded virtual Router
Our evaluation of the Router so far has focussed on external comments service models.
These topics are only part of what goes on in the internal switching unit, review batch and
forwarding. It is keenly observed that the switching function of a router actually transfers of
datagram’s from a router's incoming links to the appropriate outgoing links. In this section it
is elaborately explains about depth heat dissipation measurement of each internal unit of
router.

52
Routing processor. The routing processor executes all kind of routing protocols and
maintains the routing tables, and performs network management functions within the router.

Switching fabric. The switching fabric connects the router's input ports to its output ports.
This switching fabric and routing processor consists the following 1) Subset unit 2) Match
configuration 3) Review batch 4) Create duplicate Resolution pasted. As in the inner core
(heart of the router) includes buffer unit where it can be used to stand the packets in the queue
and undergoing Subset unit and review batch configuration.

Role of Real large buffers

Decreasing the buffer size is not only the goal to minimize power consumption because there
is the relationship between network performance and buffer resources. Buffer size and
management are directly linked to the flow control policy and QOS employed by the
network; flow control, in turn, affects network performance and resource or bandwidth
utilization. Queuing theorists are used to thinking of sizing buffers so as to prevent them from
overflowing and losing packets. When flows are desynchronized, what is the effect of large
buffers.
A single TCP flow, which experiences packet drop probability p, attains a throughput of
Roughly
x = h/RTT√p pkt/s
Here
H 0.87
RTT 142 µs

TCP connection round trip time. The network round trip time (RTT) is a finding the time
between a packet being sent and an acknowledgement being received for that packet. It is
the sum of the path latencies in the forward and reverse directions for the packet. RTT can be
affected by some of the following factors.

 Parameters of Queuing

Number of packets standing in the queue at each router between the source and the
destination.

 Design of Networks

53
Routing Topology, ie distance between routers.

The ping command is used to find this value on a IP network

P probability of packet loss. Consider a several link shared by the network where flows are
consider as r, rate of the packet moves. Consider Y= xr. Y is called total sending rate. Let as
decide Link has C is Service rate and that the buffer is large enough to keep the link fully
utilized. y > C could not fully utilized. Such cases packet drop probability p arrives. Now the
below simulation result shows that the packet loss occurs not only the reason of buffer size it
may cause due to high temperature variation.

1.6 0.040

1.4 0.035

1.2 0.030
Observed mean RTT

1.0

Standard deviation
0.025

0.8
0.020

0.6
0.015

0.4
0.010
0.2
0.005
0.0
0.000
0 200 400 600 800 1000 1200 1400 1600
Packet Size in bytes

Figure 24
In the above diagram, blue line shows that the movement of packets in the buffer unit(Y),
which causes collision inside the unit, creates heat dissipation. Standard deviation varies in
the means of collision, it is not only the reason of the small buffer size, it takes times to
verify the lookup table. Such introduces the method called Verify Lookup Time (VLT) is
the progress takes where the packets or datagram are comparing IP address to the lookup for
finding nearer node Verification time under Content Addressable Memory (CAM) which
called T3 (Single-Cycle Multiple-Logical CAM).Here it is main algorithm used in inner core
forwarding unit. Before forwarding verified in T3 CAM. Priority based packets are selected
and its logical queue depends upon the high performance network which specified live
program Audio and Video signals packets are in highest priority for 5Gbps Single clock
cycle has been proposed for each packet IP address verification in the inner core router.

54
Automatic queue management there is the high quality managing for each packet IP
address verification. As soon as the verification in progress 500 bytes packets are moving in
the forwarding unit and undergoing congestion. That path has been tested in matching table.
In power test, using a wattmeter to monitor power consumption of each core unit. The power
consumption has only a tiny increase with the core numbers growing the consumption is less
than 70w.Hence its thermal dissipation is less than 140 BTU.
There are two possibilities in throughput test, When congestion occurs
(1) Not issued any rules and issued two or more important tool rules. Packet sending
machine is adopted to conduct the performance test on system respectively by each
constructing packet of 800 bytes. The flow handled on the starting of system dropout is
considered as the limit state of the system data processing. Whether issued rules Or not, to the
first core like review batch, the system flow processing grows approximately linearly
2) Unknown packet loss locks have a great impact on performance in multithreaded.
When the packet is forwarded to the node the packet IP Address is checked with the
connected node IP address with the router. If the connected IP address is not present in the
routing table then the packets are discarded. To undergo this function the buffer takes time,
hence throughput also reduced.
Upgraded Virtual Routers(UpVR) Instead of individual and statically partitioned
First- In-First-Out (FIFO) buffers when approach this buffer work we are revisit the concept
within the confinements of the strict resource limitations of power dissipation. This is the
first attempt to incorporate a unified buffer management in flow matching table routers. The
new flavour in our endeavour stems from a fundamentally different implementation
approach: New innovation of this research is, periodic transmission of packets under silicon
based design which runs single clock operation without incurring prohibitive overhead. Most
importantly though, it enables the use of a flexible and dynamically varying thermal
management scheme, thereby replacing the conventional, static transmission and this opens a
vast research area in UpVR .

Core Design

Internal Router works at two different levels: i) Physical level(PL) ii) Logical level(LL) i)
Physical level arrangements are the electronic circuitry arrangements of the board.
Hence layer 2 identifies the forwarding mechanism which depends circuitry arrangements
also. ii) Layer 3 routing table (network layer) which logically performs the packet forwarding
function. Thus the core design concentrates PL & LL and enhancing the inner functions.
55
The forwarding table contains IP network IDs to which the node (host or router) belong.
A network ID (IP domain) is the first n bits in an IP address. The forwarding table is used
to recognize IP-packets destination for hosts that are attached to the same network as the host
itself. The forwarding table contains IP network IDs. A routing table containing IP network
Router receives an IP packet and it first extracts the destination IP address. First, it checks
whether the destination IP address is one of the IP addresses in the forwarding table. If it is
present the IP packet is sent of the corresponding network interface. Otherwise, check
whether the destination IP address is one of the IP addresses in the routing table. If it is there,
the IP packet is sent of the corresponding router. Otherwise, drop the packet or the packets
will discard. This confirmation takes time which for N no of packets stands in the queue .
Moving packets are entering into buffer unit to forward the packets to the particular routes.
Here queuing delay is the single biggest cause of uncertainty in today’s Internet as buffers
creates queuing delay and delay variance, which dissipates heat energy during this queuing
delay. It cause packet loss, and when they undergo this heat degradation it affects throughput
of entire network.

Flow verification in Routing Table:

Routers are capable enough with buffer. So Routers in built buffers flow were preceded with
following

1. Incoming packets are forwarded through MIME Block core of the control logic of EnVR.
2. Each packets are verified with RFT (Routing forwarding table).
3. It is a compact table, holding packet IP address checks with RFT. If it is matches with
existing table IP address Packets may forwarded to the hosting node and forwarded to
original connected node.
4. If it does not matches, packets are standing in the line or discarded.
5. In the true state, packets IP address is Matches with RFT and if the Packets are discarded
,network creates congestion or collision in the network.
6. This flow verification executes untill the buffer get free of packets.
1) If THERMALAvg<MINth, then no packet drops and marks occur.
2) If MINth<THERMAL Avg<MAXth, then the packets are randomly marked with a certain
probability.

56
Received the Packets

No
B<w

Yes

Drop the packet

Packets are moving to next unit

Matching No
process

IP=NN

Permit Yes Drop the packet

Packets are standing in the Queue.

Queue time Exceeds

Yes
Qt>St
=THERMA
L max>w
>B
No No

Drop the packet

Packets are forwarded to the next node

57
 .Figure 25

Flow verification in Routing Table:

Routers are capable enough with buffer. So Routers in built buffers flow were preceded with
following

1. Incoming packets are forwarded through MIME Block core of the control logic of EnVR.
2. Each packets are verified with RFT (Routing forwarding table).
3. It is a compact table, holding packet IP address checks with RFT. If it is matches with
existing table IP address Packets may forwarded to the hosting node and forwarded to
original connected node.
4. If it does not matches, packets are standing in the line or discarded.
5. In the true state, packets IP address is Matches with RFT and if the Packets are discarded
,network creates congestion or collision in the network.
6. This flow verification executes untill the buffer get free of packets.
1) If THERMALAvg<MINth, then no packet drops and marks occur.
2) If MINth<THERMAL Avg<MAXth, then the packets are randomly marked with a certain
probability.
Flow verification in Routing Table:
Routers are capable enough with buffer. So Routers in built buffers flow were preceded with
following
1. Incoming packets are forwarded through MIME Block core of the control logic of EnVR.
2. Each packet are verified with RFT (Routing forwarding table).
3. It is a compact table, holding packet IP address checks with RFT. If it is matches with
existing table IP address Packets may forwarded to the hosting node and forwarded to
original connected node.
4. If it does not match, packets are standing in the line or discarded.
5. In the true state, packets IP address is Matches with RFT and if the Packets are discarded
,network creates congestion or collision in the network.
6. This flow verification executes until the buffer get free of packets.
Proposed design called UpVR is ready to reduce the packet drop due to unstable
condition in the unit due to heavy thermal energy dissipation. Here flow matching table
works in matching progress. Queue management and its RTT movements are verified in
58
chapter 3. Whenever the queue length is arrives at either minimum threshold (Minth) and
maximum threshold the Silicon sheet implementation ( Physical based design (PBD)) logic
is maintaining in congestion problem of inner core router. The packet dropping probability
(Pd) linearly changes between zero and Pmax with the change of average queue size as well
as congestion due to Tbps transmission (Results Shown in next Chapter)
There are two main logic, drop the packets in internal or inner core router (ICR)
1) If the average queue size exceeds the size of the buffer, exceeding packets
invariably discarded or dropped.
Minth Queue size(w) is less than buffer unit (B) B > w
Maxth Queue size(w) is less than buffer unit (B) B < w
2) Logical Level of Router in forwarding table. When the packets are standing in the
queue packet IP verification of uplink and downlink creates congestion causes
pressure inside the ICR dissipates energy.
THERMALmin When packets B > w are verifying time. Denial
Thermal. THERMALmin <w <B
THERMALmax When packets B > w are verifying time. Though
Thermal leakage THERMALmax>w >B

SYSTEM IMPLEMENTATION

In the above flow chart it is clearly identified by the each step simulation process that
revealed the reason of dropping the packets. Dropping mechanism is controlled by UpVR
for which the switching and verification process heat dissipation is controlled. UpVR
sensitive to the parameters of the scheme and the dynamic traffic load . The goal of a proper
UpVR scheme is to maintain the average Dissipation thermal energy between Minth and
Maxth of the queue at one clock oscillations and in turn help avoid forced drops. It is
inappropriate for the average packet dropping probability. It has been found that with a
small average dissipation ,frequency and bandwidth in the low-load scenario, thus in order to
improve router device PBD can be used for controlling dissipation of thermal energy
20%.Here it is introduced physical based design to balance the thermal dissipation called
Silicon Oil Unbaised Page(SOUP) .

59
 In the upgraded virtual buffer size,
there are seven units were present in the function. Define Subset data is the first logic when
data’s entering to the router. Data’s are defined by IP addresses and to forward to the flow match
configuration. Here same IP address packets are segregated and call the batch match logic data.
In this unit group packet are analyzed and aligned. Aligned packets are executed and
mismatching IP Addresses packets are dropped out. During this time bottle neck occurs in buffer
unit, yellow coloured states are dissipated heat energy .in figure (ii).In order to enhance the
performance of the Buffer unit, Special design called path diversity are used in Review batch
section shown in Fig (iii). For the eastward source-destination pairs, such as S3 − D3 andS6 −
D6, the east-north (EN) turn and east-south (ES) turn are only allowed at even columns. For the
westward pairs, such as S4 − D4 and S5 − D5, the west-north (WN) turn and west-south (WS)
turn are only allowed at odd columns. Path diversity is the basic mechanism used in review
batch to reduce heat dissipation. The power profile medium thermal limits, setting suitable D is
essential to improve the thermal-limited throughput.

Table 8

Layer Dimension Thermal Conductivity λ

(W/m-K)
Heat Sink 6.0*6.0*0.7 cm3 400
DRAM Silicon 100µm 120 (Si)
DRAM Metal 2 µm 9
Proc Silicon 100 μm 120 (Si);

60
 Proc Metal 12 μm 12

Because the vertical traffic loading is redistributed by the proposed routing-based traffic
migration method, the channel loading will be unbalanced. We will propose the vertical buffer
allocation method to solve this problem, and we introduce our systematic design flow to find the
most suitable downward level. Path is cumulated into the lateral paths on XY plane and the
vertical paths along Z direction. Actually packets are moving into the proactive routing where
divided into two parts: the lateral routing and the vertical routing. UpVR design goal of the
lateral routing is to balance the lateral traffic loading on each XY plane; the Main design goal of
the vertical routing is to accommodate the various vertical traffic loadings for different thermal
limits. If the thermal limit is high, the temperature bounded traffic load is can be high. The
throughput maximization

If Thermal lim = high D = 0 can keep the lateral routing layer unchanged

If Thermal lim = low D ≥0 can keep the lateral routing layer changed.

STEP 1
A Packet arrives

STEP 2
Calculate queue waiting limit and set the average value
T=PacMax-Pacmin/2

Yes
STEP 3
T =Queue
STEP 4
limit load Calculate Thermal dissipation
(Th1), Packet dropping
probability (ᵽ1)

NO

STEP 5 Yes STEP 6

T < Queue Calculate Thermal dissipation
limit load (Th2), Packet
61 dropping probability
(ᵽ2)
 Figure 26 Detailed proposed evaluation of Router

TCP throughput has been analyzed with the buffer size impact on it. It has been found that
with increasing thermal energy are proportional to throughput and the performance of the
TCP increases.
In the proposed methodology using this path diversity method the heat dissipation is
maintained and packet dropping is slowly reduced. In path diversity silicon thin sheets are
located under the review batch to balance the temperature variation. Whenever the packets
were standing in the SRAM buffer unit its temperature raises upto 25 watts which were
calculated by watt meter. This thin film is made by silicon oil which has been also widely
used since their unusual rheological properties, low surface tension, and non-volatile. It
maintains temperature rate of router. Hence Packet dropout in the unit is reduced it can be
viewed in below simulation process.

62
 EXPERIMENTS AND RESULT
A physical based design system simulated using a Packet tracer. This system contains
seven nodes, seven nodes, one Hub and one switch where used to transfer the data from
source to destination. During simulations energy consumed and raise in temperature are
monitored continuously for each node. It is monitored that each steps includes the number
of packets transferred by all the nodes are of same length during the execution. Temperature
is measured by using of the Hotspot simulator tool. A large total of 50000 packets were
transmitted during the time of simulation throughout UpVR. Logical level is affecting
congestion; it can vary the thermal problem in physical level.
Varying queuing time conditions were taken to consideration and it creates thermal variations
in the outer core of the router. The node congestion (1) T =Queue limit load (2) T <Queue
limit load (3) T > Queue limit load.
Congestion of the node is deliberately evaluated and its simulation process shows the
temperature distribution at outer core of the router when queue limit various according to the
Standing packet congestion.

63
 1.0
T=Queue Limit
Temperature distribution
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
8 12 16 20 24 28 32 36 40 44 48 52 56 60 64
Period
100
90
Queue load limit

80
70
60
50
40

10 20 30 40 50 60
Period

Figure 27
Above simulation process output shows the outer core thermal distribution when queue limit
is equal to T = PacMax - PacMin/2.Distribution limit is not exceeds its level. Congestion is
not affecting the thermal variation at this point.

1.0
Temperature Distribution

0.8
0.6
T< Queue Limit
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
8 12 16 20 24 28 32 36 40 44 48 52 56 60 64

Period
30

25
Queue Load

20

15

10

10 20 30 40 50 60

Period

Figure 28

64
Above simulation process output shows the outer core thermal distribution remain constant,
when queue limit is T = Queue Limit. (i.e) T = PacMax - PacMin/2.Distribution limit is not
exceeds its level. Congestion is not affecting the thermal variation at this point also.
Temperature distribution

3.0 T> Queue Limit

2.5

2.0

1.5

1.0

8 12 16 20 24 28 32 36 40 44 48 52 56 60 64

260
Period
240
Queue Limit

220

200

180

160

140
10 20 30 40 50 60

Period

Figure 29
Thus the above simulation process undoubtedly shows the outer core thermal distribution
which causes drastic changes, when queue limit is T > Queue Limit. (i.e) T = PacMax -
PacMin/2.Distribution limit is gradually exceeds its level. Congestion is heavily affecting the
thermal variation at this point also.
Now Packet tracer is the software where to find no of packet dropping in the nodes. The
following figure is shown that the result of packets in the nodes in the simulation process.
Table 1 shows the parameter programme. In the simulation process four routers are connected
under one hub and this router; three routers are connected under one switch,hub and switch
are interconnected. This is the basic configuration used in simulation process.
Table 9
PARAMETER VALUE

Bandwidth 10MB

Delay 20ms

Jitter 20ms,5ms,3ms

65
Queue weight .0002

Throughput 5

Queue limit 100

Packet Size 1000 bytes

Thermal energy 3

Protocol 802.3ae

Connector type LC

Transceiver module type SFP+

Physical layer 10GBASE-R/W

Transmission rate LAN PHY—10.3125 Gbps

WAN PHY—9.95328 Gbps
Transmission Control Protocol Cubic
(TCP) Traffic Type

Figure 30 : Simulation Parameters

66
 Figure 31
This above mentioned Table is included with basic requirement of the group of network. For
ease of managing networks with minimum mentioned parameters were tabulated. Whenever
it is calculated in, TCP layer throughput is decreased according to the temperature and packet
drop increases in the network. Our simulator is used to identify packet drop mechanism
due to heat dissipation and how UpVR minimizing packet drop in the network. Packet
Movements in Downstream Upstream nodes,

1 2 6 6 7 3 4 5

Upstream nodes
6 6 1 1 1 1 3

Below output is graphical representation of simulated output taken from packet tracer.
Bandwidth , Delay, Jitter, Temperature variation, Discarding packet, average no of packets
arrival are traced out from tracer and mentioned.

67
 Packet Discard without Path diversity
50

45

40

35
Packet Ratio

30

25

20

15

10

5 no of packets discard in numbers

0
20 30 40 50 60
Temperature Distribution outer core(J)

Figure 31 (a) Parameter of simulated output with Path Diversity

In above figure when the packet arrives from nearer node the temperature of the review
batch varies from 10 to 50, Hence discarding packet due to this heat dissipation is varied
from 0 to 5.

Packet Discard with Path diversity

50

40
Packet Ratio

30

20

10

No of packet discard in numbers

0
20 30 40 50 60
Temperature Distribution in outer core(J)

Figure 33(b) Parameter of simulated output with Path Diversity

Thus the above results shown the simulation result after the implementation of UpVR.
Though the collision arrives in the router, Internal Physical based design is controlling the
heat dissipation by the means of silicon sheet arrangements of path diversity and enhancing

68
throughput of the networks inturn of avoiding no of packet loss. Discarding packet numbers
are varied from 0 to 3 no’s only.

Discussion
Although the performance of these UpVR is promising in terms of energy efficiency and
balancing thermal dissipation it is predicted in three ways.
When the packet arrives queue limit has been set it is shown in the result fig 1,2,3
1. Thermal dissipation (Th1), Packet dropping probability (ᵽ1) – Packets are not been
dropping when the buffer structure & time limit is greater than the queue limit.
2. Thermal dissipation (Th2), Packet dropping probability (ᵽ2) – Packets are not been
dropping when the buffer structure & time limit is greater than the queue limit.
3. Thermal dissipation (Th3), Packet dropping probability (ᵽ3) – Packets are been
dropping when the buffer structure & time limit is less than the queue limit.
After proposed system implemented in point no 3 (i.e) Thermal dissipation (Th3), Packet
dropping probability (ᵽ3) is reduced 10%. Hence it can be proved that our system can be used
invariably to enhance the function of networks as it reaches Quality of service in all kind of
real time applications.Energy aware QoS in wired and wireless sensor networks will ensure
guaranteed bandwidth throughout the forwarding functions.

CONCLUSION AND FUTURE RESEARCH

Optimizing performance of the node using PBD controls over energy consumption,
dissipation in data center upgrades a holistic integration of workload thermal management,
prediction, and allocation. This is best accomplished by developing a single holistic expert
system in MIME flow matching table of TCP layer protocol identification, the traffic is
reduced and its bandwidth reaches up to 4 Gbps very easily.
In this paper, it is formulated the problem of reducing packet drop in the buffer of routers as
to balance current real time applications. Our model suggests that as the number of TCP
windows as in flow based design to achieve full link utilization. This research focuses only
the bottleneck field of buffer unit and its hazards.
Here packet power inspection enhanced technology based MIME Electronic pipe line design
for avoiding delay variation in networking is basic scenario is modified. However, the
distribution of the numbers of cores in this system is limited to theoretical analysis. Proposed
Upgraded Virtual Router Stream in buffer architecture enables dynamical allocation of virtual

69
nodes and buffer slots in real-time, depending on traffic conditions. It is therefore important
to fully understand different time scales evaluation undergone in each node of throughout
sessions.The system deals for reliable transmission multimedia messages of upstream and
downstream over lossy networks.
To be successful, Path diversity mechanism is presented in buffer block review batch to
enhance network performance to reduce packet discarding. When the incoming traffic is
burst, the current internet router buffers fail to control congestion effectively. But in review
batch if we placed path diversity block in buffer unit, which produces significant
improvement in controlling packet loss due to heat dissipation without any need for special
parameterization. Commonly two experiments were done
1) Proving congestion variation in the buffer unit of router causes thermal energy
dissipation variations of the outer node using hotspot simulation tool.
2) Proposed design in buffer sections are best efforts of real time applications which
proven by evaluating packet discarding variations using Packet tracer simulator. It
explores and the outcome shows that path diversity implementation has reduced
packet loss when compared to traditional mechanisms. The simulation is to maintain
adaptive buffer space under a sudden change in congestion level which prevents the
heavy energy leakage in the inner core when congestion occurs.
The system deals for reliable transmission multimedia messages of upstream and downstream
over lossy networks.
As we have many research approaches to save energy and to control its dissipation energy.
It satisfies the key problem.1) UpVR architecture is the most thermal controlling efficient
design. 2) It decreases overall 10% of thermal dissipation and enhances performance of node.
Therefore, for future research to successfully optimize performance and energy consumption
in data centers, it must provide a holistic integration of workload prediction, allocation, and
thermal management using PBD

70