Storage Mag Online Sept Updated 92010

Managing the information that drives the enterprise
STORAGE Vol. 9 No. 6 September 2010
Virtualizing
NAS
File storage is getting out
of hand and most NAS systems
can’t scale to accommodate the
growth. It’s time to consider
virtualizing your NAS storage.
P. 13
ALSO INSIDE
5 ILM is back!
8 Gotta have primary storage dedupe
21 Time for a network tune-up
27 Readers rate their midrange arrays
33 Backup for the 21st century
36 Just say no to storage stacks?
38 RAID is still first line of defense
1
STORAGE september 2010
R E G I O NA L S O L U TI O N P RO V I D E R
STORAGE 5
inside | september 2010
Editorial: ILM Lives Again!

EDITORIAL Information lifecycle management faded into
oblivion without getting serious notice. But it’s back with a
new name and more realistic goals. by RICH CASTAGNA
Primary Storage Dedupe: Requisite for the Future

8 STORWARS Few companies are immune to the runaway
growth of file data. While automated tiering and thin provisioning
can help users cope with capacity demands, more drastic
measures, like primary storage data reduction, are needed.
by TONY ASARO
Virtualizing NAS
13 Traditional file storage typically lacks the scalability most
companies require and results in disconnected islands of
storage. File virtualization can pool those strained resources
and provide for future growth. by JACOB GSOEDL
Top 10 Tips for Tuning Your Storage Network

21 Before you blame your storage system for poor performance,
take a look at your storage network. These 10 tips will help
you find and fix the bottlenecks in your storage network
infrastructure. by GEORGE CRUMP
Quality Awards V: Compellent Regains Top

Midrange Arrays Spot
27 Loyal, and apparently very satisfied users,
propelled Compellent to the front of the pack
among midrange storage systems. Read all of the
results of our fifth Quality Awards service and reliability
survey for midrange arrays. by RICH CASTAGNA
Getting in Front of Backup

33 HOT SPOTS Adding disk to the process revolutionized backup,
but now the focus is shifting from back-end hardware to
newer front-end technologies, like CDP, replication, source-side
deduplication and snapshots, that are poised to provide even
greater backup efficiencies. by LAUREN WHITEHOUSE
Storage Vendors are Stacking the Deck

36 READ/WRITE Storage vendors have been busy creating
server-to-application product stacks. It looks like the type
of ploy that will give them more leverage and take it away
from you. by ARUN TANEJA
RAID Still Rules for First-Line Defense

38 SNAPSHOT RAID has taken some knocks lately, like criticism
that it’s a nearly 30-year-old technology that can’t stand up to
the rigors of a modern data storage environment. But 96% of
the respondents to our survey said they rely on some form of
RAID. by RICH CASTAGNA
Vendor Resources
39 Useful links from our advertisers.
3 Storage September 2010 Cover illustration by ENRICO VARRASSO
3PAR Inc. | 4209 Technology Drive, Fremont, CA 94538
510.413.5999 | www.3PAR.com
storage stacks? editorial | rich castagna
Just say no to
ILM lives again!
t
Information lifecycle management faded into oblivion
without getting serious notice. But it’s back with a
new name and more realistic goals.
Midrange arrays
Quality Awards:
HE PHRASE “information lifecycle management” seemed to serve as a cure

for insomnia when it was first introduced. Even its acronym—ILM—failed to
catch on in an industry that loves acronyms. And saying “ILM” to a storage
manager produced glazed eyes, a stony silence or both.
But the concept of moving data to the most appropriate type of storage
based on its current usefulness (or age) still sounds like an idea worth
waking up to, doesn’t it? Everybody’s swimming upstream against a rising
tide of data with fewer and fewer dollars to
keep them afloat, so why wouldn’t you want
to ensure that you’re not blowing bucks on
Saying “ILM” in
network tune-up
expensive storage for data with little or no public is still a

Time for a
value?
Most shops do care and are taking a hard
no-no, but whatever
look at where they put their data. You don’t you call it—storage
hear a lot of “ILM” chatter but, hey, that’s
exactly what it is. When the idea of ILM rolled
tiering or simply
around to open systems—hijacked from the smart storage
mainframe world’s hierarchical storage man-
agement (HSM)—more people seemed to be
management—
hung up on determining the value of the data it’s back.
Virtualizing NAS
rather than its ideal location.

As a result, data classification became a new catch phrase, and a hand-
ful of companies with classification technologies sprung up. The premise
was that you needed to know more about a piece of data than when it was
created, when it was changed last or how big it was. All of that can be use-
ful information, but you need to have some real intelligence about that
data if you have any hope of determining its proper disposition.
People say the more you know, the better. So why not crack open your
data files and see what’s inside? After all, you can’t tell if data should hang
around on premium platters or be shelved to some near-line system or
Primary storage
the equivalent of storage Siberia if you don’t know its true worth. But to
data dedupe
know all that, you would need to get your business units involved, which
is about the time ILM gets laid to rest.
But you can’t keep a good idea down, and ILM is back and being taken
more seriously than ever. Saying “ILM” in public is still a no-no, but whatever
you call it—storage tiering or simply smart storage management—it’s back.
What’s different this time is that we’re focused on the problem. We’re looking
5 Copyright 2010, TechTarget. No part of this publication may be transmitted or reproduced in any form, or by any means, without permission in writing from
Storage May 2010
the publisher. For permissions or reprint information, please contact Mike Kelly, VP and Group Publisher (mkelly@techtarget.com).
STORAGE
storage stacks?
Just say no to
at location, the placement of data, much more closely. We’ve essentially stopped
looking for a perfect solution long enough to consider what might be good enough
or at least expedient.
But that explanation is a little too simple; ILM is back because we have more
choices about where to put things than we did before. Solid-state storage might
be the key catalyst for ILM’s renewal. When solid state began to trickle into
enterprise storage systems, the debate was over
how to determine what applications, if any, were
Not every company
Midrange arrays
Quality Awards:
worth the incredibly high price of flash. Solid-

staters said forget $/GB and think in terms of dollarshas the app or the
per I/O, which added a new dimension to the
argument. Eventually, someone realized that bucks to add a
rather than parking data on solid-state storage, pretty expensive
we should just let it hang out there for as long
as it’s needed. tier at the top of the
So the idea of moving data dynamically and storage triangle, but
automatically came into play. Forget about crack-
ing files or indexing content; let’s just see how the same principles
could be used to
network tune-up
often and how fast the data is needed. Not every

Time for a
company has the app or the bucks to add a pretty

expensive tier at the top of the storage triangle, move data around
but the same principles could be used to move from, say, SAS to
data around from, say, SAS to SATA. There might
not be sophisticated data classification going on SATA.
behind the curtain, but it’s a practical solution.
Cloud storage, being taken more and more seriously by enterprises every day,
tosses yet another tier into the mix. And clever startups like StorSimple and
Nasuni have built appliances that almost seamlessly integrate the cloud with
data center storage.
Virtualizing NAS
And now that LTO-5 is here, tape is suddenly cool again. LTO-5’s 3 TB capacity
and 240 MBps throughput (both with compression) definitely reinforce tape’s
status as a bona fide storage tier.
If your storage vendor doesn’t offer some form of automated data movement,
ask when it will. Just as thin provisioning is already entrenched in most enter-
prise storage systems, and the way data reduction is moving along that same
route, automated tiering will become a basic part of a storage vendor’s system
management set. If it isn’t, then you might want to consider another vendor. 2
Rich Castagna (rcastagna@storagemagazine.com) is editorial director of the Storage

Primary storage
data dedupe
Media Group.
* Click here for a sneak peek at what’s coming up in the October 2010 issue.
6 Storage September 2010

ENERATION
CKUP
TS NOW
It’s about disk. It’s about networks. It’s about time. EMC is the leader in disk-based backup and recovery.
Learn more now. www.EMC.com/products/category/backup-recovery.htm
EMC2, EMC, and where information lives are registered trademarks or trademarks of EMC Corporation in the United States and other countries. © Copyright 2010 EMC Corporation. All rights reserved.
storage stacks? StorWars | tony asaro
Just say no to
Primary storage dedupe:

Requisite for the future
Tools like automated tiering and thin provisioning
t
help users cope with capacity demands; but more drastic
Midrange arrays
Quality Awards:
measures, like primary storage data reduction, are needed.
EN YEARS AGO, 10 TB was considered a large storage environment. Now

it’s common to have hundreds of terabytes and there are even environ-
ments with petabytes of storage in the double-digit range. It’s safe to
assume that data storage capacity growth will continue over the next 10
years as storage environments measured in exabytes begin to emerge
network tune-up
and, over time, become mainstream. I actually talked to one customer

Time for a
who claimed they would have an exabyte of data in the next three years.
Having that much physical storage in the data center is ultimately un-
tenable. So how do we solve the problem? A big part of the answer will be
provided through a number of technologies. Hard disk drives will continue
to become denser. Higher capacity disk drives
have the ability to store more data within the Having that much
same given physical space. However, fatter disk
drives impact application performance. There- physical storage
fore, intelligent tiering that enables demotion in the data center
Virtualizing NAS
and promotion of active and inactive data

between fast and dense storage tiers will is ultimately
balance performance and capacity. untenable.
Storage optimization technologies such as
thin provisioning provide a better way to use the capacity you already have
within your storage systems. Storage systems that use traditional provi-
sioning methods typically have 50% to 70% of their capacity allocated but
unused. Users who implement thin provisioning have a much higher utiliza-
tion rate. If you can reduce allocated but unused capacity to 20%, it will
yield significant savings in a petabyte world. For an environment with 1 PB
Primary storage
of storage, implementing thin provisioning could result in 300 TB to 500 TB

data dedupe
of capacity being saved. If you have 10 PB, then we’re talking a savings of
3 PB to 5 PB.
These are great leaps, and I submit that another major leap will be data
reduction (data deduplication) for primary storage. The math is simple and
the value proposition is a no-brainer. Even moderate dedupe is economi-
cally attractive. If your data is consuming 100 TB of disk space and you’re

STORAGE
storage stacks?
Just say no to
able to cut that in half, you would reclaim 50 TB of capacity. That’s a fairly
modest 2:1 ratio, which should be easily achievable. If you were able to
get a 5:1 ratio, you’re talking approximately 80 TB of reclaimed capacity.
If we consider a petabyte data center, you can save 500 TB on the conser-
vative side (a 2:1 reduction ratio), and 800 TB if you’re more optimistic (5:1
ratio). For 10 PB of data, the result could be a capacity savings of up to 8 PB.
The savings are staggering when you consider just the capital costs,
Midrange arrays
Quality Awards:
but it also drives down your maintenance costs. When you factor in
the impact on operations and people
resources, the value proposition becomes The savings are
even more compelling. And if you add all
of that to power, cooling and floor space
staggering when
savings, primary dedupe can completely you consider
change the IT landscape.
You would expect every storage system
just the capital
vendor to have deployed primary deduplica- costs, but it also
tion by now, but there are some significant
drives down your
network tune-up
issues to overcome, such as:

Time for a
• There’s a potential performance impact, maintenance costs.

which is a no-no in storage.
• Primary dedupe may require more internal resources (e.g., memory
and CPU). In some cases, it isn’t simply just a question of adding
more because of design limitations.
• Even if there are no physical resource issues, some storage systems
may require architectural changes to support deduplication. This
could take years or may not even be possible in some cases.
• Regardless of what anyone tells you, primary dedupe is complex
Virtualizing NAS
technology that’s typically not a core competency for most vendors.

• If something goes wrong, the risk—losing data forever—is high, so
vendors are cautious.
There are two storage system vendors that provide primary dedupe
today. While both vendors have modest adoption, it certainly isn’t exten-
sive. The reason for this is that their deduplication products have distinct
limitations in terms of scalability and performance. However, we’re on the
threshold of more and better products coming to market. You’ll see an-
nouncements later this year and in 2011, and it will grow from there.
Primary storage
Data dedupe is a form of virtualization and I believe it will become as

data dedupe
ubiquitous as server virtualization within all tiers of data storage. The

amount of storage we have today and the growth over the next decade
is a pervasive problem that has to be solved. The reality is that we can’t
just keep throwing money at the problem. 2
Tony Asaro is senior analyst and founder of Voices of IT (www.VoicesofIT.com).

Weather The Storm With
AdvizeX Technologies
Don’t let the storm of factors forcing you to evaluate
the viability of your storage infrastructure
leave you stranded
AdvizeX has the technical information and know how you need.
Creative IT Solutions
3 to 1 Technical to Sales Staff Contact us today
No Cost Assessments Available for your no fee
assessment
800.366.6096 | www.advizex.com
STORAGE
storage stacks?
Just say no to
STORAGE
Integrating Cloud and
Traditional Backup Apps
COMING IN OCTOBER
10 Tips: Managing
Storage for Virtual
What Storage
Managers Are Buying
Servers and Desktops
Midrange arrays
Quality Awards:
Cloud storage services are Over the last eight years,

making some inroads into Storage magazine and
Virtualization on the server
enterprise storage opera- SearchStorage.com have
and desktop side of IT
tions, especially for backup fielded the twice-yearly
shops has provided a rela-
and disaster recovery (DR). Storage Purchasing
tively easy way to reduce
As standalone services, Intentions survey to deter-
physical systems. But the
cloud backup and cloud mine the purchasing plans
technology has introduced
DR can be useful, but for of storage professionals for
problems for storage
many firms, they would be the current and upcoming
managers who need to
far more attractive alterna- years. This article reports
effectively configure their
tives if they could be inte- on and analyzes the results
resources to meet the
grated with more traditional
network tune-up
needs of a consolidated of the latest edition of the

backup methods. We’ll look survey, and provides insight
infrastructure. We offer
Time for a
at how some backup and into emerging trends. Find

the top practical tips for
DR app vendors are working out what technologies your
managing storage in virtu-
to integrate cloud storage peers are interested in for
alized server and desktop
services. the coming year.
environments.
And don’t miss our monthly columns and commentary,

or the results of our Snapshot reader survey.
Virtualizing NAS
TechTarget Storage Media Group
STORAGE
Vice President of Editorial Site Editor Site Editor
Mark Schlack Ellen O’Brien Susan Troy
Editorial Director Senior News Director

Rich Castagna Dave Raffo Site Editor
Site Editor Sue Troy
Senior Managing Editor Senior News Writer Andrew Burton
Kim Hefner Sonia Lelii UK Bureau Chief
Managing Editor Antony Adshead
Senior Editor Features Writer Heather Darcy
Primary storage
Ellen O’Brien Carol Sliwa

data dedupe
Features Writer
Creative Director Senior Managing Editor Todd Erickson TechTarget Conferences
Maureen Joyce Kim Hefner
Executive Editor and Director of Editorial Events
Contributing Editors Associate Site Editor Independent Backup Expert Lindsay Jeanloz
Tony Asaro Megan Kellett W. Curtis Preston
James Damoulakis Editorial Events Associate
Steve Duplessie Editorial Assistant Jacquelyn Hinds
Jacob Gsoedl David Schneider
Storage magazine
Storage magazine 275 Grove Street
Subscriptions: Newton, MA 02466
www.SearchStorage.com editor@storagemagazine.com

All LTO storage tapes are the same
Alll LTO storage tapes arr tha same
Awll LTOW strgee tayps r tha sayme
Alll strgg lapees hare da sssaymez
A!! sstg p lape zz h@r da same zz
@l!l *!467 b$©kup t%yp£ arr® saym
wllzz eLtho...b......ddfe7....................
.............................................................
.............................................................
.............................................................
....... Sure, all LTO storage tapes are
the same. Except when they fail.
Your entire enterprise depends on the integrity of your data. Choose Sony LTO
Ultrium™ 5 cartridges, backed by our 60 years of magnetic tape expertise.
The LTO specifications don’t require our A3MP magnetic particles, high-tech
chemical binder and high-strength base film. But your data deserves no less.
Visit sony.com/LTO
© 2010 Sony Electronics Inc. All rights reserved. Reproduction in whole or in part without written permission is prohibited. Features and specifications are subject to change without notice.
Sony, the Sony logo and make.believe are trademarks of Sony. LTO, the LTO logo, Ultrium and the Ultrium logo are trademarks of HP, IBM and Quantum.
Virtualizing
storage stacks?
Just say no to
NAS
Midrange arrays
Quality Awards:
Companies of all sizes

are being inundated with
unstructured data that’s
straining the limits of
network tune-up
traditional file storage.

Time for a
File virtualization can

pool those strained
resources and provide
for future growth.
By Jacob Gsoedl
Virtualizing NAS
u NSTRUCTURED DATA is growing at an unprecedented rate in all industries and

Primary storage
has become one of the top challenges for IT departments. Market data
data dedupe
from a variety of analyst and research firms shows a congruent picture:

In most companies, the amount of unstructured data (file based) out-
strips structured data; it’s spread across the enterprise; and it tends to
reside on a motley assortment of isolated file stores that range from file
servers to network-attached storage (NAS). Management pain points have
reached a critical level and associated costs are skyrocketing.

STORAGE
storage stacks?
Just say no to
How we ended up in this dilemma is well understood. On the one

hand, we have the simplicity of implementing unstructured data stores
via Windows and Linux file servers with directly attached and storage-
area network (SAN) storage; on the other hand, we have traditional
NAS systems that are based on scale-up architectures with inherent
limitations to scale. For instance, until NetApp released Ontap 8 it
lacked advanced clustering
and a global namespace; the
Midrange arrays
Quality Awards:
only way to extend beyond NAS virtualization

a single NetApp filer was to
buy a larger filer or deploy
another one running inde-
terminology
pendently from already Namespace is the organization and presenta-
installed systems. tion of file-system data, such as directory
The data storage industry structure and files.
is keenly aware of the situa- In a non-shared namespace, file-system
tion, and vendors have taken information is confined to a single physical
different approaches to pro- machine and not shared with others. Traditional
network tune-up
vide file system and NAS scale-up NAS and server-based file stores
Time for a
virtualization products that are examples of products with non-shared

help to overcome the chal- namespaces.
lenge at hand. Even though Conversely, a shared namespace, also
progress has been made, referred to as a global namespace, combines
adoption has been tepid. “It the namespace of multiple physical machines
has taken almost 10 years for or nodes into a single namespace. It can be
block-based virtualization to implemented by aggregating the namespaces
take place,” said Greg Schulz, of multiple machines and presenting them as
founder and senior analyst a single federated namespace, as is usually
Virtualizing NAS
at Stillwater, Minn.-based the case in file-system virtualization and

StorageIO Group. “NAS virtual- clustered NAS products, or it can be achieved
ization is still in an early via clustered file systems where a single file
stage and it will take time system spreads across multiple physical nodes.
for it to be widely adopted.” Scale-up NAS is a file-based storage
system that scales by replacing hardware with
FOUR WAYS TO faster components, such as faster CPUs, more
VIRTUALIZE FILE ACCESS memory and more disks. Its namespace spans
Virtualizing file access by one or two nodes clustered for high availability.
putting a logical layer be- Scale-out NAS is a file-based storage
Primary storage
tween back-end file stores system that provides scaling by adding nodes
data dedupe
and clients, and providing a to the cluster. Available in N+1 (single redundant
global namespace is clearly node) or N+M (each node has a redundant node)
the most promising approach high-availability configurations, they provide a
to tackling the unstructured namespace that spans multiple nodes, allow-
data challenge. It’s akin to ing access to data throughout all nodes in the
block-based storage virtual- namespace.

STORAGE
storage stacks?
Just say no to
ization, however, there isn’t a single method of implementing file-access

virtualization. Instead, we have several architectural approaches competing
for a potentially lucrative file-access virtualization market.
File-system virtualization (aggregation) is one way of
1. virtualizing file access. At a high level, file-system virtualization

accumulates individual file systems into a pool that’s accessed
by clients as a single unit. In other words, clients see a single large name-
space without being aware of the underlying file stores. The underlying
Midrange arrays
Quality Awards:
file store could be a single NAS, or a

mesh of various file servers and NAS One of the great
systems. File-system virtualization
products address two main problems: benefits of file-system
They give users a single virtual file store; virtualization is that
and they offer storage management
capabilities such as nondisruptive data it can be deployed in
migration and file-path persistency existing environments
while files are moved between different
physical file stores. without having to rip
network tune-up
One of the great benefits of file- out existing servers

Time for a
system virtualization is that it can be

deployed in existing environments with- and NAS storage.
out having to rip out existing servers and
NAS storage. On the downside, file-system aggregation doesn’t address
the problem of having to manage each file store individually.
Clustered file systems are another way of virtualizing file
2. access. Clustered file systems are part of next-generation

NAS systems designed to overcome the limitations of tradi-
tional scale-up NAS. They’re usually composed of block-based storage
Virtualizing NAS
nodes, typically starting with three nodes and scaling to petabytes of file
storage by simply adding additional nodes. The clustered file system
glues the nodes together by presenting a single file system with a single
global namespace to clients. Among the vendors offering NAS systems
based on clustered file systems are FalconStor Software Inc.’s HyperFS,
Hewlett-Packard (HP) Co.’s StorageWorks X9000 Network Storage Systems,
IBM’s Scale Out Network Attached Storage (SONAS), Isilon Systems Inc.,
Oracle Corp.’s Sun Storage 7000 Unified Series, Panasas Inc., Quantum
Corp.’s StorNext and Symantec Corp.’s FileStore.
Clustered NAS is a third way of virtualizing file access. Clus-
3.
Primary storage
tered NAS architectures share many of the benefits of clus-

data dedupe
tered file-system-based NAS. Instead of running a single file

system that spreads across all nodes, clustered NAS systems run com-
plete file systems on each node, aggregating them under a single root
and presenting them as a single global namespace to connected clients.
In a sense, clustered NAS is a combination of a scale-out, multi-node
storage architecture and file-system aggregation. Instead of aggregating

STORAGE
storage stacks?
Just say no to
OPEN SOURCE NAS VIRTUALIZATION

NAS virtualization products are also available as open source software. For instance,
the Apache Hadoop Distributed File System (HDFS) handles distribution and
redundancy of files, and enables logical files that far exceed the size of any one
data storage device. HDFS is designed for commodity hardware and supports any-
where from a few nodes to thousands of nodes. Another example of an open source
file system is the Gluster clustered file system for building a scalable NAS with a
Midrange arrays
Quality Awards:
single global namespace.

Instead of spending a lot of money for traditional NAS systems, an open source
file system running on inexpensive hardware components seems like a good alter-
native. But open source file systems are usually not a good choice for the enter-
prise. They require significant tuning and maintenance efforts, as well as experts
intimately familiar with the intricacies of the chosen software, and they don’t come
with the support that traditional NAS vendors offer. Availability, reliability, perform-
ance and support come first for enterprise storage, and these attributes are diffi-
cult to achieve with open source software. Open source file systems are a great
choice for cloud storage providers and companies that have to make money on
network tune-up
storage, as well as for the research and educational sector, but they’re usually
Time for a
not the product of choice in the enterprise.
file systems of heterogeneous file stores, they aggregate file systems

on native storage nodes. The BlueArc Corp. Titan and Mercury series of
scale-out NAS systems are prime examples of clustered NAS systems.
NAS gateways can also be viewed as file-system virtualiza-
4. tion devices. Sitting in front of block-based storage, they

Virtualizing NAS
provide NFS and CIFS access to the block-based storage they

front end. Offered by most NAS vendors, they usually allow bringing
third-party, block-based storage into the NAS and, if supported by the
NAS vendor, into the global namespace.
NAS systems and gateways based on clustered file system or clus-
tered NAS architectures are next-generation NAS systems and won’t
integrate with existing legacy file stores; they usually replace them or
run in parallel with them. This makes them more difficult to deploy as
well as more expensive than file-system virtualization products. However,
the benefit of having to manage a single NAS, rather than many small
Primary storage
data silos that are simply aggregated by a file-system virtualization

data dedupe
product into a single namespace, more often than not justifies the
additional effort and cost.
FILE-SYSTEM VIRTUALIZATION USE CASES AND SELECTION CRITERIA

Because ripping out existing file stores and replacing them with a
scale-out NAS isn’t an option in many situations, file-system virtual-

STORAGE
storage stacks?
Just say no to
ization products that aggregate the various file stores into a single
global namespace can be viewed as complementary to scale-out and
traditional NAS systems, especially during the extended time of transi-
tioning from legacy file stores. “Many customers buy a NAS to get fea-
tures like replication, archiving and snapshots, but they don’t require
these for all files,” said Brian Gladstein,
vice president (VP) of marketing at Even in companies
AutoVirt Inc. “We give them the ability
that can centralize
Midrange arrays
Quality Awards:
to mix existing low-end file stores with

fast filers and provide them with a single their unstructured
namespace.”
Even in companies that can central- data onto a NAS with
ize their unstructured data onto a NAS global namespace
with global namespace support, there
will likely always be some storage silos support, there will
that live outside the NAS. It could likely always be some
be departmental data or data that’s
deemed unworthy to reside on com- storage silos that live
network tune-up
paratively expensive NAS storage. File- outside the NAS.

Time for a
system virtualization products allow

combining rogue file stores with NAS devices into a global namespace.
A second use case for file-system aggregation is data migration. Acqui-
sitions, storage infrastructure upgrades and data relocation projects
are among the reasons for migrating data from one physical location
to another. Because file-system aggregation products virtualize access
to heterogeneous file stores, they’re also simple yet effective data
migration solutions. Another use case for file-system aggregation is
automated storage tiering. Equipped with policy engines for defining
Virtualizing NAS
data migration rules based on file-system metadata—such as last

access date, file size and file type—they enable automatic data move-
ment to suitable storage tiers based on defined policies.
File-system virtualization products are available as appliances and
software-only products. A software-only product offers the benefits
of more flexible deployment on hardware of your choice, and the products
usually have a lower degree of vendor lock-in. Conversely, appliance-
based file-system virtualization products come in a proven, perform-
ance-optimized package and, because hardware and software are
provided by the same vendor, there’s less risk of finger pointing.
Primary storage
When comparing file-system virtualization products, the level at which

data dedupe
virtualization occurs is a relevant evaluation criteria. For instance, while

Microsoft Distributed File System (DFS) provides share-level virtualization,
a product like F5 Networks Inc.’s ARX series provides file-level virtualization.
Intrusiveness and ease of deployment are also relevant character-
istics to consider during a product evaluation. Ideally, a file-system
virtualization product should require minimal client changes and the

STORAGE
storage stacks?
Just say no to
virtualized data on the back-end file stores shouldn’t be changed.

File-system support must also be considered. While some systems
support only CIFS, products like F5’s ARX and EMC Corp.’s Rainfinity
support CIFS and NFS, which is relevant in environments with both
Windows and Linux file stores. The presence of a policy engine and
its capabilities are critical if the product’s intended use is for
data mobility and automated storage tiering.
Midrange arrays
Quality Awards:
FILE-SYSTEM VIRTUALIZATION PRODUCT SAMPLER

File-system virtualization products are offered by a number of vendors,
each coming from a different background with varying objectives.
AutoVirt File Virtualization software: Like Microsoft DFS,
AutoVirt is a software-only product that runs on Windows servers.
The AutoVirt global namespace uses the CIFS protocol to interact
with file servers, clients and DNS. When a client requests a file, DNS
facilitates resolution to the appropriate storage device. The global name-
space acts as an intermediary between client and DNS. With the AutoVirt
global namespace in place, client shortcuts refer to the namespace. The
network tune-up
namespace is the authority on the location of networked files and provides

Time for a
the final storage referral with the help of DNS.

AutoVirt can be introduced nondisruptively to clients, without the
need to make any changes on clients, by populating the AutoVirt
namespace server with the shares of existing file stores. Although it
can be done manually, a data discovery service automates discovery
of existing file stores and populates the AutoVirt namespace server
with metadata. This differs from Microsoft DFS, which requires clients
to be configured with the new DFS shares, rather than continuing to
use existing file shares.
Virtualizing NAS
Also contrary to Microsoft DFS, AutoVirt provides a policy engine that

enables rule-based data mobility across the environment to migrate,
consolidate, replicate and tier data without affecting end-user access
to networked files. Currently available for CIFS, AutoVirt plans to release
a version for NFS by year’s end.
EMC Rainfinity file virtualization appliance: Rainfinity is a family
of file-system virtualization products that virtualize access to unstruc-
tured data, and provide data mobility and file tiering services. The Rain-
finity Global Namespace Appliance provides a single mount point for
clients and applications; the Rainfinity File Management Appliance
Primary storage
delivers policy-based management to automate relocation of files to

data dedupe
different storage tiers; and the Rainfinity File Virtualization Appliance

provides nondisruptive data movement.
Contrary to F5’s ARX, the Rainfinity File Virtualization Appliance archi-
tecture is designed to switch between in-band and out-of-band opera-
tions as needed. The appliance is out-of-band most of the time, and data
flows between client systems and back-end file stores directly. It sits

STORAGE
storage stacks?
Just say no to
outside the data path until a migration is required and then switches
to in-band operation.
F5 ARX Series: Acquired from Acopia in 2007 and rebranded as F5
ARX, the F5 ARX series is an inline file-system virtualization appliance.
Usually deployed as an active-passive cluster, its located between
CIFS/NFS clients and heterogeneous CIFS/NFS file stores, presenting
virtualized CIFS and NFS file systems to clients. Unstructured data is
presented in a global virtualized namespace. Built like a network
Midrange arrays
Quality Awards:
switch, it’s available with 2 Gbps ports (ARX500), 12 Gbps ports

(ARX2000) and 12 Gbps ports plus two 10 Gbps ports (ARX4000).
With a focus on data mobility and
storage tiering, F5’s ARX comes with With a focus on data
strong data mobility and automated
storage tiering features. Orchestrated mobility and storage
by a policy engine, it performs bidirec- tiering, F5’s ARX
tional data movements between differ-
ent tiers of heterogeneous storage in comes with strong
real-time and transparently to users. data mobility and
network tune-up
Similar to AutoVirt, policies are based

automated storage
Time for a
on file metadata, such as last-accessed

date, creation date, file size and file tiering features.
type.
The fact that F5 ARX is an appliance allows it to provide a performance-
optimized product that’s hard to match by a software-only solution.
Built on a split-path architecture, it has both a data path that passes
data straight through the device for tasks that don’t involve policies,
and a control path for anything that requires policies. “We are a DFS
on steroids,” said Renny Shen, product marketing manager at F5.
Virtualizing NAS
“While DFS gives you share-level virtualization, we give you file-level

virtualization.”
Microsoft DFS: Microsoft DFS is a set of client and server services
that allow an organization using Microsoft Windows servers to organize
distributed CIFS file shares into a distributed file system. DFS provides
location transparency and redundancy to improve data availability in
case of failure or heavy load by allowing shares in multiple locations
to be logically grouped under a single DFS root.
DFS supports the replication of data between servers using File
Replication Service (FRS) in server versions up to Windows Server
Primary storage
2003, and DFS Replication (DFSR) in Server 2003 R2, Server 2008 and
data dedupe
later versions.
Microsoft DFS supports only Windows CIFS shares and has no
provision for bringing NFS or NAS shares into the DFS global namespace.
Furthermore, it lacks a policy engine that would enable intelligent data
movements. As part of Windows Server, it’s free and a good option for
companies whose file stores reside mainly on Windows servers.

STORAGE
storage stacks?
Just say no to
FILE VIRTUALIZATION OUTLOOK

Access to unstructured data hasn’t changed much in the past 15 years,
but big changes are happening now. NAS system architectures are
moving toward more scalable, multi-node scale-out architectures
with global namespace support. NAS behemoth NetApp finally incor-
porating technology acquired from Spinnaker in its Ontap 8 release,
enabling customers to build multi-node NetApp clusters, is indicative
of the change.
Midrange arrays
Quality Awards:
File-system virtualization products are complementing traditional

scale-up and next-generation scale-out NAS systems to provide a global
namespace across heterogeneous file stores in the enterprise. While
they’re currently mostly deployed for the purpose of data mobility and
storage tiering, they’re likely to play a significant role in the future in
providing an enterprise-wide, global unified namespace for all
unstructured data. 2
Jacob Gsoedl is a freelance writer and a corporate director for business

systems. He can be reached at jgsoedl@yahoo.com.
network tune-up
Time for a
Virtualizing NAS
Primary storage
data dedupe

storage stacks?
Just say no to
Midrange arrays
Quality Awards:
TOP 10 TIPS FOR

TUNING YOUR
STORAGE NETWORK
network tune-up
Time for a
Storage performance issues are often not related

to the storage system at all, but rather to the storage
e
network that links servers to disk arrays. These
10 tips will help you find and fix the bottlenecks in
your storage network infrastructure. By George Crump
Virtualizing NAS
VERY SO OFTEN there’s a moment of calm in a data storage manager’s life

where nothing is broken and there aren’t any fires to put out. As rarely as
these times might occur, the momentary calm should be taken advantage
of rather than savored. This is your opportunity to get some of the kinks out
of your storage network so you can eliminate the next emergency before
it happens or just be better prepared when it does. We spoke with experts
from storage networking vendors—Brocade, Cisco, Emulex and Virtual In-
struments—to discuss what storage managers should do to prepare their
storage networks for the future and to maximize their investments.
Primary storage
The first few tips that follow have more to do with being prepared than
data dedupe
actually tinkering with your storage-area network (SAN), but all of our experts
agreed that trying to fine-tune a SAN without adequate preparation is like
driving down a freeway without headlights. Before you can roll up your
sleeves and get under the hood, you have to do some preparation. The rest
of our tips go into more detail, describing specific steps (often at no cost)
that you can take to improve SAN performance, efficiency and resiliency.

STORAGE
storage stacks?
Just say no to
Tip 1. Know what you have

The No. 1 recommendation in fine-tuning your storage network is to first
know what you have in the environment. If you have a problem and need
to bring in your vendor’s tech experts, the first thing they’re going to
want is an inventory of your networking environment. If you do the
inventory ahead of time, you’ll likely pay less for any necessary profes-
sional services and it may even help you avoid having to engage them
Midrange arrays
Quality Awards:
in the first place.

It’s important to document each
host bus adapter (HBA), cable and
It’s important to document
switch in the environment while each host bus adapter
noting how they’re interconnected.
You should also record the speeds
(HBA), cable and switch
they’re actually set at, and the in the environment while
versions of the software or drivers
they’re running. While all of this
noting how they’re
may seem painfully obvious, an interconnected.
network tune-up
inventory of what the storage

Time for a
network consists of and how it’s configured is the type of document

that can quickly fall off the priority list during the urgencies of a typical
IT workweek. Taking time to level set and understand what’s in the
environment, and how it has changed, is critical.
Documenting this information may even pinpoint some areas that
are ripe for fine-tuning. We’ve seen cases where over the course of time
users have upgraded to 4 Gb Fibre Channel (FC) and, for some reason,
their inter-switch links (ISLs) were still set at 1 Gb. A simple change to
the switch configurations effectively doubled their performance. If they
Virtualizing NAS
hadn’t taken the time to do an inventory, this obvious mistake may never
have come to light.
This could be a zero-cost tip because the information can be captured
and stored in spreadsheets. While manually keeping track of this informa-
tion is possible, in today’s rapidly changing, dynamic data center it’s be-
coming a less practical approach. Storage environments change fast and
IT staffs are typically stretched thin, so manually maintaining an infra-
structure isn’t realistic. Vendors we spoke to, and many others, have soft-
ware and hardware tools that can capture this information automatically.
Of course, those tools aren’t free or as cheap as a spreadsheet. But
Primary storage
if you weigh their cost against the cost of manually capturing the data,
data dedupe
or the cost of missing an important change to the network environment,

it can be a good investment. Automated storage resource management
(SRM) tools also vary in the data they capture and the level at which
they capture it. Many simply poll devices and record status data, while
others tap the physical layer and analyze network frames.

STORAGE
storage stacks?
Just say no to
Tip 2. Know what’s going on

After you’ve developed a good picture of the components in your storage
network infrastructure, the next step is to fully understand what those
devices are doing at a particular moment in time. Many switch and HBA
vendors build some of these capabilities into their products. But instead
of going to each device to see its view of traffic conditions, it may be
better to find a tool that can provide consolidated real-time feedback
Midrange arrays
Quality Awards:
on how data is traversing your network. There are software solutions

and physical layer access tools that
can report on the infrastructure traffic. There are software
The tools that can monitor network
devices specifically are important
solutions and physical
because, as all of our experts pointed layer access tools that
out, there are situations where oper-
ating systems or applications report
can report on the
inaccurate information when com- infrastructure traffic.
pared to what the device is reporting.
network tune-up
These tools can be used for trend analysis and, in some cases, they
Time for a
can simulate an upcoming occurrence of a data storage infrastructure

problem. For example, if an ISL is seeing a steady increase in traffic
(see Tip 6), the ability to trend that traffic growth will help identify how
soon an application rebalance or an increase in ISL bandwidth will be
required. Other tools will report on CRC or packet errors to ports, which
can indicate an upcoming SFP failure.
Tip 3. Know what you want to do

With your inventory complete and good visibility into your SAN estab-
Virtualizing NAS
lished, the next step is to figure out what network changes will provide
the most benefit to the organization. You may have discovered SAN
features that need to be enabled, or perhaps you have new applications
or an accelerated rollout of current initiatives that need to be planned.
Knowing how activities such as those will impact the rest of the envi-
ronment and what role the storage infrastructure has to play in those
tasks is critical. Generally, the goals come down to increasing reliability
or performance, but they may also be to reduce costs.
Tip 4. Limit the impact

Primary storage
When you feel you’re at the stage where you’re ready to make changes
data dedupe
to the environment, the next step is to limit the sphere of impact as

much as possible by subdividing the SAN into virtual SANs (VSANs).
Subdividing (in a worst-case scenario) changes made to the environ-
ment that yield unexpected results, like preventing a server from
accessing storage or even causing an outage, will have limited reper-
cussions across the infrastructure. Limiting the sphere of impact is by

STORAGE
storage stacks?
Just say no to
itself an important fine-tuning step that will help create an environment

that’s more resilient to changes in the future, and can help contain
problems. For example, an application may suddenly need an excessive
amount of storage resources; subdividing the SAN will help contain it
and keep the rest of the infrastructure from being starved. This aspect
of fine-tuning shouldn’t require any new purchases as it’s a setup and
configuration process.
Midrange arrays
Quality Awards:
Tip 5. Test to learn, learn to test

Although it may seem to be something of a luxury, one key to fine-tuning
is to have a permanent testing lab that can be used to try out proposed
changes to the environment or to simulate failed conditions. Lab testing
lets you explore the alternatives and develop remedies without impacting
the production network. In speaking with our experts, and in our own
experience, most SAN emergencies result from implementing a new
feature in the storage array or on the SAN. If you lack the resources
to create a lab environment, an alternative may be to work with your
network tune-up
infrastructure vendors, as many have facilities that can be used to

Time for a
recreate problems or to test the implementation of new features.

Storage I/O performance is typically high on a fine-tuning top 10 list,
and although it didn’t make it into our top five tips, it rounds out the
rest of the list. Before performance issues are tackled, it’s important
that the environment be documented, understood and made as resilient
as possible. While slow response time due to lack of performance tuning
is a concern, zero response time because of poor planning is a lot worse.
Tip 6. Understand how you’re using ISLs

Virtualizing NAS
ISLs (interconnects between switches) are critical areas for tuning, and
as a storage-area network grows, they become increasingly important
to performance. The art of fine-tuning an ISL is often an area where
different vendors will have conflicting opinions on what a good rule
of thumb is for switch fan-in configurations and the number of hops
between switches. The reality is that the latency between switch con-
nections compared to the latency of mechanical hard drives is dramati-
cally lower, even negligible; however, in high fan-in situations or where
there are a lot of hops (servers crossing multiple switches to access
data), ISLs play an important role.
Primary storage
The top concern is to ensure that ISLs are configured at the correct
data dedupe
bandwidth between the switches, which seems to be a surprisingly

common mistake as mentioned earlier. Beyond that, it’s important to
measure the traffic flow between hosts and switches, and the ISL traffic
between the switches themselves. Switch reporting tools will provide
much of this information but, as indicated earlier, a visual tool that
measures switch intercommunication may be preferable.

STORAGE
storage stacks?
Just say no to
Based on the traffic measurements, a determination can be made

to rebalance traffic flow by adjusting which primary switch the server
connects with, which will involve physical rewiring and potential server
downtime. Another option is to add ISLs, which increases bandwidth but
consumes ports and, to some extent, further adds to the complexity of
the storage architecture.
Midrange arrays
Quality Awards:
Tip 7. Use NPIV for virtual machines

Server virtualization has changed just about everything when configur-
ing SANs and one of the biggest challenges is to identify which virtual
machines are demanding the most from the infrastructure. Before server
virtualization, a single server had a single application and communicated
to the SAN through a single HBA;
now virtual hosts may have many Server virtualization
servers trying to communicate
with the storage infrastructure all has changed just about
through the same HBA. It’s critical everything when configuring
network tune-up
to be able to identify the virtual

SANs and one of the biggest
Time for a
machines that need storage I/O

performance the most so that they challenges is to identify
can be balanced across the hosts,
instead of consuming all the re- which virtual machines
sources of a single host. N_Port ID are demanding the most
Virtualization (NPIV) is a feature
supported by some HBAs that lets from the infrastructure.
you assign each individual virtual
machine a virtual World Wide Name (WWN) that will stay associated
Virtualizing NAS
with it, even through virtual machine migrations from host to host. With
NPIV, you can use your switches’ statistics to identify the most active
virtual machines from the point of view of storage and allocate them
appropriately across the hosts in the environment.
Tip 8. Know thy HBA queue depth

HBA queue depth is the number of pending storage I/Os that are sent
to the data storage infrastructure. When installing an HBA, most storage
administrators simply use the default settings for the card, but the default
HBA queue depth setting is typically too high. This can cause storage
Primary storage
ports to become congested, leading to application performance issues.

data dedupe
If queue depth is set too low, the ports and the SAN infrastructure itself
aren’t used efficiently. When a storage system isn’t loaded with enough
pending I/Os, it doesn’t get the opportunity to use its cache; if essen-
tially everything expires out of cache before it can be accessed, the
majority of accesses will then be coming from disk. Most HBAs set the
default queue depth between 32 to 256, but the optimal range is actually

STORAGE
storage stacks?
Just say no to
closer to 2 to 8. Most initiators can report on the number of pending

requests in their queues at any given time, which allows you to strike
a balance between too much and not enough queue depth.
Tip 9. Multipath verification

Multipath verification involves ensuring that I/O traffic has been
distributed across redundant paths. In many environments, our experts
Midrange arrays
Quality Awards:
said they found multipathing isn’t working at all or that the load isn’t
balanced across the available paths. For example, if you have one path
carrying 80% of its capacity and the other path only 3%, it can affect
availability if an HBA or its connection fails, or it can impact application
performance. The goal should be to ensure that traffic is balanced fairly
evenly across all available HBA ports and ISLs.
You can use switch reports for multipath verification. To do this, run
a report with the port WWNs, the port name and the MBps sorted by
the port name combined with a filter for an attached device type equal
to “server.” This is a quick way to identify which links have balanced
network tune-up
multipaths, which ones are currently acting as active/passive and

Time for a
which ones don’t have an active redundant HBA.
Tip 10. Improve replication and backup

performance
While some environments have critical concerns over the performance
of a database application, almost all of them need to decrease the
amount of time it takes to perform backups or replication functions.
Both of these processes are challenged by rapidly growing data sets
that need to be replicated across relatively narrow bandwidth connec-
Virtualizing NAS
tions and ever-shrinking backup windows. They’re also the most likely
processes to put a continuous load across multiple segments within
the SAN infrastructure. The backup server is the most likely candidate
to receive data that has to hop across switches or zones to get to it.
All of the above tips apply doubly to backup performance. Also con-
sider adding extra HBAs to the backup server and have ports routed to
specific switches within the environment to minimize ISL traffic. 2
George Crump is president and founder of Storage Switzerland, an IT analyst

firm focused on the storage and virtualization segments.
Primary storage
data dedupe

storage stacks?
Just say no to
QUALITY AWARDS V:
Compellent regains
top midrange arrays spot
Midrange arrays
Quality Awards:
The fifth edition of our service and reliability

survey for midrange arrays shows that users of midrange
storage systems are pretty darned satisfied with their purchases.
By Rich Castagna
network tune-up
Time for a
MIDRANGE STORAGE ARRAY vendors seem to be

doing a lot of things right. On our latest Quality
Awards survey of midrange arrays, all of the
finalists scored so well that it’s hard to find
fault with any of their product lines. This is
the fifth time we’ve canvassed users about
the service and reliability of their midrange
storage arrays, and these users showed the
highest level of satisfaction with this product
Virtualizing NAS
class to date. With a record overall score of 7.12

for midrange systems, Compellent Technologies
Inc. has regained the crown that Dell Inc. snagged
in our previous survey.
Compellent’s win was earned with a consistent
performance, garnering the top scores in all five
of our rating categories—sales-force compe-
tence, initial product quality, product features,
product reliability and technical support. In
Primary storage
winning all of the categories, the company’s

data dedupe
Storage Center products picked up

ratings greater than 7.00 in four of
them; the ratings range from
1.00 to 8.00.

STORAGE
storage stacks?
Just say no to
MIDRANGE MIGHT
But even with such an excellent showing, Compellent must still share
at least of little of the Quality Awards spotlight. While not seriously
challenging Compellent’s overall score of 7.12, all eight finalists finished
with overall scores higher than 6.00—the first time that has happened
for midrange arrays and a rarity for any Quality Awards survey.
Hewlett-Packard (HP) Co. rode its EVA and P4000 lines to a strong
Midrange arrays
Quality Awards:
second-place finish, scoring an overall 6.73, with Hitachi Data Systems

just a shade behind at 6.67. But a strong 1-2-3 finish isn’t the end of the
story, as the rest of the group was bunched closely behind the leaders
with ratings ranging from NetApp’s 6.59 to Oracle Corp.’s (Sun) 6.38. Five
of the eight vendors had scores higher than the winning overall score in
our last survey.
SHARPER SALES TEAMS

On most Quality Awards surveys, the lowest user ratings typically appear
in the sales-force competence category, as was the case on the most
network tune-up
recent midrange array survey. With tighter budgets and often urgent
Time for a
needs, storage managers expect sales reps and their support teams
to be responsive and well-informed. Just a few years ago on our second
midrange survey, the overall average for the sales category was a tepid
5.28, indicating that users’ expectations were likely met but rarely
exceeded. This time around, the category average is 6.47, suggesting
strong—and probably effective—
efforts by vendors’ sales forces.
Compellent picked up its first
ABOUT THE SURVEY category win with a 6.81, buoyed
Virtualizing NAS
by scores of 7.00-plus for the

The Storage magazine/SearchStorage.com
Quality Awards are designed to identify and statements “My sales rep keeps
recognize products that have proven their my interests foremost” and “The
quality and reliability in actual use. The vendor’s sales support team is
results are derived from a survey of quali- knowledgeable.” In all, Compellent
fied readers who assess products in five came out on top for four of the
main categories: sales-force competence,
six category statements. HP didn’t
initial product quality, product features,
product reliability and technical support. lead for any single statement, but
Our methodology incorporates statistically lined up ratings running from 6.45
valid polling that eliminates market share to 6.96 to finish second with a
Primary storage
as a factor. Indeed, our objective is to iden- category average of 6.67. Third-

data dedupe
tify the most reliable products on the mar- place finisher Hitachi (6.59) was
ket regardless of vendor name, reputation
high scorer for two of the category
or size. Products are rated on a scale of 1.00
to 8.00, where 8.00 is the best score. A total statements (“My sales rep is
of 315 respondents provided 497 midrange knowledgeable about my industry”
storage array evaluations. and “My sales rep understands my
business”).

STORAGE
storage stacks?
Just say no to
Whether it’s in response to the rigors of a down economy or reaction to

surveys such as this, vendors appear to be redoubling their sales efforts.
IMPLEMENTATION AND INITIAL QUALITY

A positive sales experience is a great way to start a relationship, but
it can quickly be derailed if problems arise during deployment of the
product. Here, too, in the initial product quality category, our midrange
Midrange arrays
Quality Awards:
array vendors seem to be surpassing expectations as well as their

previous performances. Ratings
were high across the board in this
category, with Compellent posting
PRODUCTS IN THE SURVEY above-7.00 scores for five of the
The following vendors and midrange array six statements en route to a 7.21
model lines were included in the Quality
Awards survey. The number of responses
category average finish. Compel-
for the finalists are included in parentheses lent’s only sub-7.00 score was a
after the product names. 6.93 for “This product was installed
without any defects.” Hitachi’s
network tune-up
• 3PAR InServ E200 or F200/F400*

rating of 6.94 for that statement
Time for a
• Atrato Inc. Velocity 1000 (V1000)* topped Compellent by the slimmest

• BlueArc Corp. Titan 2000/3000 Series, possible margin.
Mercury* HP nudged out Dell for the second
• Compellent Technologies Inc. Storage spot in the category (6.88 vs. 6.80),
Center (14) and garnered the only other 7.00-
• DataDirect Networks Inc. S2A Series* plus score (a 7.06) for the key
• Dell Inc. CX Series or Dell EqualLogic statement “This product is easy
PS Series (47) to use.” Not too far behind Dell,
• EMC Corp. Clariion CX Series (110) the rest of the field was packed
Virtualizing NAS
• Fujitsu Eternus DX400 Series* almost too tightly to see any

• Hewlett-Packard (HP) Co. StorageWorks space among them; a mere .05
EVA Series and P4000 Series (76) points separated Oracle (Sun),
• Hitachi Data Systems Universal Storage Hitachi and IBM (tied), NetApp
Platform (USP) VM or AMS Series (35) and EMC Corp.
• IBM DS4000/DS5000/DS6000 (74) And if perfection is the goal
• NetApp FAS200/FAS900/FAS2000 (69) of midrange array vendors, they
appear to be well on their way,
• NEC Corp. D3/D4/D8 Series*
as the average score for all eight
• Oracle Corp. Sun 6000/7000 Series (35) products for the statement “This
Primary storage
• Pillar Data Systems Axiom 300/500/600* product was installed without any
data dedupe
• SGI InfiniteStorage 4000/5000/6000 defects” was a glossy 6.81.

Series*
• Xiotech Corp. Magnitude 3D or FEATURES AND FUNCTIONS
Emprise* A midrange storage system from
* Received too few responses to be included among just a few years ago would hardly be
finalists.
recognizable by today’s standards.

STORAGE
storage stacks?
Just say no to
Responding to user demands, vendors have tricked out midrange arrays

with the kinds of features and capabilities that you could only once get
with enterprise-class storage systems. Improved features—and more of
them—equate to contented users as evidenced by record high scores in
the product features rating category.
Compellent’s 7.26 average score easily outdistanced NetApp, which
notched a very solid 6.75 giving it a small margin over third-place finisher
Midrange arrays
Quality Awards:
HP (6.68) which it turned nudged out IBM (6.61) by the same margin.
Compellent’s Storage Center scored the highest on all seven category
statements in the product features category, ranging from a 7.07 (“This
product’s snapshot features meet my needs”) to a 7.36 (“This product’s
remote replication features meet my needs”). By delivering these
“bread-and-butter” features along with its signature Fluid Data auto-
mated tiering, Compellent may be raising the bar a bit for all midrange
systems vendors.
That’s not to suggest that any of the product lines are slackers
when it comes to features. The overall average for the category was a
network tune-up
6.64, the highest we’ve seen and substantially higher than the previous
Time for a
mark of 6.33. The average scores for key midrange array requirements
were high for all eight products, such as a 6.79 for “This product’s
capacity scales to meet my needs,” highlighted by Hitachi’s 7.11 (the
only other 7.00-plus score in the category) and Compellent’s 7.31.
STORAGE YOU CAN COUNT ON

THE HEAVY LIFTERS The true test of a storage system is
how well it performs after the show-
room shine wears off. Each of the eight
Vendor/Product Average installed capacity
Virtualizing NAS
(TB) vendors’ product lines passed the test

on this survey easily, with an across-
Hitachi Data Systems USP 87
VM or AMS Series the-board 6.74 average in the product
reliability category; this is not only the
EMC Clariion CX Series 70
highest average ever for that category,
IBM 66 but the highest average of any category
DS4000/DS5000/DS6000
to date.
Oracle Sun 6000 or 60 Compellent again netted the highest
7000 Series
ratings for each of the five category
NetApp 59 statements, ranging from 7.14 to 7.36
FAS200/FAS900/FAS2000
Primary storage
and rolling up to a 7.27 average, its

data dedupe
Hewlett-Packard EVA Series 53 highest category score on the survey.

and P4000 Series
But it wasn’t alone in “seven heaven.”
Dell CX Series or Dell 43 Second-seeded Hitachi earned three
EqualLogic PS Series
7.00-plus scores for the statements
Compellent Storage Center 41 related to meeting service-level re-
quirements, having very little downtime

A
LIT Y WARD
UA
S
Q
MIDRANGE ARRAYS
TO
IN
E
S
RAG
E MAGAZ
n Compellent Storage Center

storage stacks?
Just say no to
n Dell CX Series or
OVERALL RATINGS
Dell EqualLogic PS Series
n EMC Clariion CX Series
Compellent
HP
n Hewlett-Packard StorageWorks HDS
EVA Series and P4000 Series
NetApp
n Hitachi Data Systems USP VM Dell
or AMS Series
IBM
n IBM DS4000/DS5000/DS6000 EMC
n NetApp FAS200/FAS900/FAS2000
Midrange arrays
Quality Awards:
Oracle
n Oracle Sun 6000/7000 Series

4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50
SALES-FORCE COMPETENCE INITIAL PRODUCT QUALITY

Compellent Compellent
HP HP
HDS Dell
NetApp Oracle
IBM HDS
network tune-up
Dell IBM
Time for a
EMC NetApp
Oracle EMC
4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50
PRODUCT FEATURES PRODUCT RELIABILITY

NetApp HDS
HP HP
Virtualizing NAS
IBM EMC
HDS Dell
Oracle NetApp
Dell IBM
EMC Oracle
4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50
TECHNICAL SUPPORT WOULD YOU BUY THIS PRODUCT AGAIN?

HDS NetApp
Primary storage
data dedupe
HP Dell
Dell EMC
EMC HP
IBM HDS
NetApp IBM
Oracle Oracle
4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 65% 70 75 80 85 90 95 100
*% Yes
Based on a 1.00-8.00 scoring scale

STORAGE
storage stacks?
Just say no to
and needing very few unplanned patches, on the way to an impressive

6.97 category average. Third-place HP joined Compellent and Hitachi in
the plus-7.00 club with a 7.03 for the “unplanned patches” statement
and finished with an average rating of 6.72, which put it just .01 ahead
of EMC’s score of 6.71.
CONTINUING SUPPORT FOR THE PRODUCT

Past Quality Award surveys for midrange arrays had scores in the
Midrange arrays
Quality Awards:
technical support category that hovered around the lows seen in the
sales-force competence category. This survey isn’t any different, with
support getting the second-lowest overall category average. But the
twist here is that the score is still fairly high at 6.59, led once again by
Compellent (7.02). Hitachi (6.71) racked up its second second-place fin-
ish, with HP (6.69) hard on its heels with another strong performance.
HP finished second or third in all five ratings categories.
The only statement in the support category that Compellent didn’t
score top marks on was “Vendor’s third-party partners are knowledge-
able”; instead, HP and Oracle (Sun) tied for the lead with a score of 6.72.
network tune-up
It’s a significant mark for those two vendors, as 54% of HP respondents

Time for a
and 37% of Oracle respondents said they purchased their systems from
VARs.
Midrange vendors are also delivering on their support promises. One of
Compellent’s two 7.29 category scores was for “Vendor supplies support
as contractually specified,” a statement that all vendors scored well on
for a group average of 6.79 (high in the category). Well-trained support
staffs were also recognized on the survey, with Compellent (7.07), HP
(6.87) and Hitachi (6.80) all standing out for the statement “Support
personnel are knowledgeable.”
Virtualizing NAS
DO IT AGAIN
In addition to the specific statements in each rating category, we
asked survey respondents a more subjective question: All things con-
sidered, would you buy the product again? Over our five surveys for
midrange arrays, the responses have been generally positive and very
steady, with an average of 77% to 79% saying “Yes” across all product
lines. This time, the “buy again” numbers jumped, reflecting the higher
category ratings and, undoubtedly, greater satisfaction with the entire
class of midrange storage products.
Primary storage
Overall, 89% of respondents said they would take the plunge again with
data dedupe
the same product, led by Compellent’s eerily perfect 100%, NetApp and Dell
both at 94% and the rest of the field ranging from EMC’s 87% to Oracle’s
(Sun) 83%. Not too shabby when it comes to satisfied customers. 2
Rich Castagna (rcastagna@storagemagazine.com) is editorial director of the

Storage Media Group.

storage stacks?
hot spots | lauren whitehouse
Just say no to
Getting in front of backup

Learn about a handful of key technologies that can
t
help storage managers meet their backup recovery
time objectives (RTOs) by making the first steps—
data capture and transfer—simpler and more efficient.
Midrange arrays
Quality Awards:
HE FOCUS ON backup modernization during the last few years has been
squarely on the backup target device: tapes and disks. That’s where the
majority of users have made the most changes. But now that so many
users and IT shops have become disk friendly, there’s a new focus on
the front end of the backup process: the capture and transfer phase.
In 2004, nearly 60% of Enterprise
Storage Group (ESG) survey respondents Now that so many
reported backing up directly to tape. By
network tune-up
2010, only 20% were using tape exclu- users and IT shops
Time for a
sively. These days, approximately 80% have become disk

of IT organizations tell ESG they’re aug-
menting backup processes with disk, friendly, there’s a
which helps them meet backup windows new focus on the front
and recovery time objectives (RTOs).
Still, exponential data growth means end of the backup
greater backup demands and a need process: the capture
for new backup processes. As a result,
technologies such as continuous data and transfer phase.
Virtualizing NAS
protection (CDP), replication, source-side

deduplication and snapshot are being implemented more frequently. ESG
research found a significant uptake in several of these technologies:
while the use of snapshots grew only 2% between 2008 and 2010, replication
use increased 34%, CDP expanded by 58% and deduplication use improved
66% in the same two-year period.
SNAPSHOT AND IMAGE-LEVEL BACKUP

What if you could eliminate your backup window, accelerate system recovery
and facilitate efficient disaster recovery (DR)? Effectively, that’s what snap-
Primary storage
shot- and image-based backup can deliver. A snapshot is a copy of a volume

data dedupe
or file system created at a specific point in time. Taking advantage of snap-

shot functionality for backup can dramatically reduce the impact on apps
by eliminating the backup window, providing RTOs of seconds to minutes and
enabling better recovery point objectives (RPOs) by enabling more frequent
copies per day.
Image backup uses snapshot technology to create a point-in-time

STORAGE
storage stacks?
Just say no to
image of a system (such as hardware configuration, OS, applications and

data), storing it in a single portable file. Because the recovery point is cap-
tured “hot,” critical applications don’t have to be shut down during backup.
This approach eliminates the backup window and enables rapid whole-system
recovery to any system (virtual or physical), including to dissimilar hardware.
Both of these methods are efficient in the capture, transfer and storage of
data. After the initial base copy is made, only incremental blocks are captured
Midrange arrays
Quality Awards:
and stored.
CDP
CDP technology continuously captures changes to data at a file, block or
application level, supporting very granular data capture and recovery options.
It time stamps each write and mirrors it to a continuous data protection
retention log. When a recovery is needed, the CDP engine creates an image
of the volume for the point in time requested without disrupting the production
application.
Block-level CDP operates at the logical volume level and records every
network tune-up
write. This type of continuous data protection stands out at transparent data
Time for a
capture and presentation of views at different points in time. Typically

running on the same server as the application it’s protecting, file-level
CDP operates at the file-system level and records any changes to the file
system. Application-aware CDP tracks critical application process points
within the CDP data stream that can greatly simplify recovery, such as
transaction-consistent database checkpoints or application-consistent
points within email applications.
Continuous data protection completely eliminates discrete backups,
replacing them with a transparent, continuous data capture process that
Virtualizing NAS
puts very low overhead on production servers. Because it captures data as

it’s created, that data is immediately recoverable. This allows CDP-based
solutions to deliver near-zero RPOs.
REPLICATION
Replication is the bedrock of these strategies and it’s increasingly being
used for data protection as a standalone process to provide operational and
disaster recovery for applications with tight RPOs or RTOs; as a method of
consolidating distributed data for centralized file-level backup; or in con-
junction with snapshot or CDP to maintain an off-site copy and facilitate
Primary storage
disaster recovery. Replication provides an exact mirror copy of data on a

data dedupe
local or remote primary system that can be mounted to rapidly recover

from a failure. Storage capacity and bandwidth are optimized with block-
level updates and network compression after the initial copy is made.
Replication is available on host systems, storage arrays or in network-
based products. Typically, array- and network-based products replicate at
the block level and host-based offerings replicate at the file-system level.

STORAGE
storage stacks?
Just say no to
Host-based replication operates asynchronously, while array- and network-

based replication are configurable for synchronous or asynchronous modes.
Synchronous replication occurs in real-time as data is written to primary
storage; then it’s replicated on secondary storage. Asynchronous replication
occurs in near real-time. Once data has been completely written to primary
storage, the written data is replicated on secondary storage.
Midrange arrays
Quality Awards:
SOURCE-SIDE DEDUPLICATION
Deduplication identifies and eliminates redundancy, storing only unique
data and shortcuts to unique data for duplicates. Data deduplication’s role
in optimizing backup processes is fairly well documented; however, the
focus has mostly been on target-side deduplication solutions. Source-side
deduplication ensures that only changed segments are backed up after the
initial full copy. That means significantly less data is captured, transferred
and stored on disk. This reduces the time needed to perform backups. Be-
cause the backup window requirements are minimal, it’s possible to back
up more frequently, which increases the number of recovery points on disk
network tune-up
storage to meet RPO and RTO requirements.

Time for a
A wholesale replacement of file-level backup is likely for many organiza-

tions today, according to ESG research. For example, 55% of IT organizations
surveyed by ESG plan to replace existing file-level backup with snapshot
and/or CDP solutions. That said, the integration of snapshot, replication,
CDP and deduplication into existing backup platforms to augment file-level
approaches seems to be a strong trend. That’s why several backup vendors
have made recent strides to match capture techniques to recovery objective
policies, simplifying implementations and optimizing the front end of backup
processes.2
Virtualizing NAS
Lauren Whitehouse is a senior analyst focusing on backup and recovery soft-

ware and replication solutions at Enterprise Strategy Group, Milford, Mass.
Primary storage
data dedupe

storage stacks?
read/write | arun taneja
Just say no to
Storage vendors
stacking the deck
t
Storage vendors have been busy creating
server-to-application product stacks. It looks
like the type of ploy that will give them
Midrange arrays
Quality Awards:
more leverage, and take it away from you.
HERE’S A FUNDAMENTAL SHIFT of titanic proportions taking place in IT. No, I don’t
mean the massive shift toward using disk in favor of tape to protect data.
I’m also not referring to the fundamental changes occurring in storage ar-
chitectures to improve its interaction with virtual server technologies nor
the increased usage of solid-state storage or automated storage tiering.
What’s causing this big shift is the crazed passion with which the industry
network tune-up
seems to be heading into building proprietary stacks from the server all the
Time for a
way to the application.

Cisco Systems Inc., for example, is building servers and partnering with
VMware Inc. and EMC Corp. to create what EMC calls a Virtual Computing
Environment (VCE) solution. In reaction to Cisco getting into the server
business, Hewlett-Packard (HP) Co. mainstreamed its ProCurve networking
group within the company and bought 3Com Corp. to ensure it had a strong
networking alternative to Cisco. And Oracle Corp. purchased Sun Microsys-
tems Inc. and then made known its intention to build a complete vertical
stack that will be tightly integrated (read proprietary) and use its own
Virtualizing NAS
virtualization technology.
Even Hitachi Data Systems, seemingly content being a best-of-breed
high-end and midrange storage supplier, felt it needed to do something. It
reached back to its parent company and announced its own vertical stack
using Hitachi servers, which will have a special console to integrate the
stack. NetApp then went on to do its deal with Cisco and VMware as a
counterpoint to EMC’s moves.
Storage vendors are scurrying to line up partners so they aren’t left out.
The question is if any of this craziness is necessary or warranted. My answer
is a flat “No.” I have the advantage of having seen the minicomputer revolu-
Primary storage
tion, then the PC revolution followed by the client/server revolution. Now

data dedupe
we’re undergoing a virtual “everything” revolution. I still remember a previ-

ous “vertical” stack era when users chose partners for life. If you belonged
to the IBM camp you lived and died by it. Ditto for Burroughs or Honeywell
or Unisys. Device interoperability didn’t exist, and applications often worked
on only one stack. You effectively belonged to the computer company.
For decades, the industry worked very hard to break this “own every-

STORAGE
storage stacks?
Just say no to
thing” mentality. We developed standard interfaces (such as SCSI) to make

storage work with many different systems. EMC did a phenomenal job to
create a “best-of-breed” storage solution that worked with any system
that supported a SCSI interface. Then other storage interfaces, like Fibre
Channel (FC), were developed; it took a few years to get the interoperability
kinks out, but it got done. The benefit to IT has been immeasurable and has
happened across all disciplines. Applications can run on many different
Midrange arrays
Quality Awards:
OSes, APIs are available for managing devices and printers work with every
system in the market. In addition, TCP/IP opened up a new world. We’ve finally
arrived at an era where choice matters, where best of breed matters. You still
place your bet on a vendor, but not for everything.
Now it seems we’re heading back to the ’70s. It doesn’t matter who
started the “stack war” or who’s partnering with whom. What matters is
that your choices are about to be taken off the table. For example, keep an
eye on Oracle over the next decade;
they now control hardware, database, It doesn’t matter who
storage and server virtualization.
network tune-up
The vendors’ reasons are clear, started the “stack war”

Time for a
but the direction is all wrong. As a or who’s partnering with

customer, you want to be able to buy
EMC storage even if you have all Sun whom. What matters is
servers. Or you may want NetApp that your choices are about
systems for NAS and EMC for SANs.
Maybe 3PAR’s your choice for virtual to be taken off the table.
server storage and you favor Nexsan
Corp. for data archiving. Best-of-breed systems keep everyone on their toes,
and I’d hate to see that disappear as each player opts to partner with others.
Virtualizing NAS
I can understand how the vertical stack strategy is in the best interest
of Cisco or Oracle. What I don’t see is why vendors such as EMC, Hitachi
Data Systems, NetApp and VMware would want to play this game. Their
success was built on delivering best-of-breed products and being able to
play with everyone. So why limit yourself by choosing partners?
You will be the final arbiter. You’ll either let the big guys dictate what you’ll
buy or you won’t. It might seem innocuous right now, but it does matter.
I like choices. I like that VMware has Microsoft Corp. and Citrix Systems
Inc. to compete with, and that 3PAR, EMC, Hitachi Data Systems, IBM and
NetApp are contenders for high-end storage.
Primary storage
Ultimately, you’ll vote with your dollars. Don’t forget: It was users who
data dedupe
threw out the proprietary stacks a few decades ago. You have the same kind
of leverage now, but at an earlier stage in the process. It’s up to you. 2
Arun Taneja is founder and president of the Taneja Group, an analyst and
consulting group focused on storage and storage-centric server technologies.
He can be reached at arunt@tanejagroup.com.

snapshot
storage stacks?
Just say no to
Aging RAID still an effective data protection

technology
RAID HAS TAKEN some knocks lately, like criticism that it’s a nearly 30-year-old tech-
nology that can’t stand up to the rigors of a modern data storage environment. But
maybe it’s been around so long because it’s so good: 96% of respondents to our
survey rely on some form of RAID. The most-used RAID configuration isn’t much of a
surprise, as 87% use RAID 5, followed by RAID 1 (52%) and RAID 10 (40%). Seventy-five
percent of RAID users employ more than one type on RAID on their storage systems,
Midrange arrays
Quality Awards:
and nearly 20% juggle four different RAID configurations in their shops. But that’s not
to suggest users are totally enamored with RAID, as their two biggest gripes are in-
efficient use of disk capacity (36%) and lengthy rebuild times (32%); however, 10% of
respondents didn’t see any particular shortcomings. RAID appears to be doing its job
well: 72% had to perform RAID rebuilds at least once in the last year and although re-
builds took a little while (54% said three hours to 12 hours), 93% reported that they
didn’t lose any data. To quote one respondent: “RAID rocks!” —Rich Castagna
How do you determine On average, how long did your

network tune-up
which RAID level to use? RAID rebuilds take to complete?

Time for a
5% Other Less than two hours 20%
8% Use storage Three hours to 12 hours

system vendor’s 54%
recommendations
63% 13 hours to 24 hours 18%
9% Based on Based on
disk capacity specific
application 25 hours to 48 hours 4%
15% Use same needs
RAID level on all
systems Longer than 48 hours 3%
Virtualizing NAS
0% 10 20 30 40 50 60
%
Rate the following data protection technologies
45
in order of their importance to your company.
(Least important = 1.0, Most important = 5.0)
Rating Protection technology

4.2 RAID
3.9 Traditional backup
2.8 Continuous data protection
Have had a second drive
Primary storage
2.8 Replication
fail in the same RAID
data dedupe
1.3 Cloud backup services

group during a rebuild.
“ We spin about 500 TB of array storage and have yet to

experience negative issues in our environment that
can be attributed to RAID devices.” —Survey respondent

STORAGE
storage stacks?
Just say no to
Check out the following resources from our sponsors:
3PAR, page 4
Use 10 Best Practices to Improve Your Storage Management
Midrange arrays
Quality Awards:
VMware vSphere with 3PAR Utility Storage: Simple and Efficient VMware vSphere Deployment with 3PAR InServ
Storage Servers
AdvizeX Technologies, page 10

network tune-up
EMC Backup and Recovery Solutions, page 7

Time for a
E-Guide: Best Practices for Data Protection and Recovery in Virtual Environments
E-Guide: How Dedupe and Virtualization Work Together

Virtualizing NAS
Primary storage
data dedupe

Storage Mag Online Sept Updated 92010

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Storage Mag Online Sept Updated 92010

Загружено:

Авторское право:

Доступные форматы

Managing the information that drives the enterprise

STORAGE Vol. 9 No. 6 September 2010

Editorial: ILM Lives Again!

Primary Storage Dedupe: Requisite for the Future

Top 10 Tips for Tuning Your Storage Network

Quality Awards V: Compellent Regains Top

Getting in Front of Backup

Storage Vendors are Stacking the Deck

RAID Still Rules for First-Line Defense

ILM lives again!

HE PHRASE “information lifecycle management” seemed to serve as a cure

expensive storage for data with little or no public is still a

rather than its ideal location.

worth the incredibly high price of flash. Solid-

often and how fast the data is needed. Not every

company has the app or the bucks to add a pretty

Rich Castagna (rcastagna@storagemagazine.com) is editorial director of the Storage

6 Storage September 2010

Primary storage dedupe:

measures, like primary storage data reduction, are needed.

EN YEARS AGO, 10 TB was considered a large storage environment. Now

and, over time, become mainstream. I actually talked to one customer

and promotion of active and inactive data

of storage, implementing thin provisioning could result in 300 TB to 500 TB

8 Storage September 2010

issues to overcome, such as:

• There’s a potential performance impact, maintenance costs.

technology that’s typically not a core competency for most vendors.

Data dedupe is a form of virtualization and I believe it will become as

ubiquitous as server virtualization within all tiers of data storage. The

Tony Asaro is senior analyst and founder of Voices of IT (www.VoicesofIT.com).

9 Storage September 2010

Cloud storage services are Over the last eight years,

needs of a consolidated of the latest edition of the

at how some backup and into emerging trends. Find

And don’t miss our monthly columns and commentary,

TechTarget Storage Media Group

Editorial Director Senior News Director

Ellen O’Brien Carol Sliwa

11 Storage September 2010

Companies of all sizes

traditional file storage.

File virtualization can

u NSTRUCTURED DATA is growing at an unprecedented rate in all industries and

from a variety of analyst and research firms shows a congruent picture:

13 Storage September 2010

How we ended up in this dilemma is well understood. On the one

only way to extend beyond NAS virtualization

virtualization products that are examples of products with non-shared

at Stillwater, Minn.-based the case in file-system virtualization and

14 Storage September 2010

ization, however, there isn’t a single method of implementing file-access

1. virtualizing file access. At a high level, file-system virtualization

file store could be a single NAS, or a

One of the great benefits of file- out existing servers

system virtualization is that it can be

2. access. Clustered file systems are part of next-generation

tered NAS architectures share many of the benefits of clus-

tered file-system-based NAS. Instead of running a single file

15 Storage September 2010

OPEN SOURCE NAS VIRTUALIZATION

single global namespace.

not the product of choice in the enterprise.