Академический Документы
Профессиональный Документы
Культура Документы
SDN Abstractions
Ma
p
Map
Reduc
e
Reduc
e
HDFS
HDFS
Map
Ma
p
Map
Reduc
e
Reduc
e
Determine
paths with
minimal
congestion
Install New
paths
Draw Backs
Demand gather at network is
ineffective
Assumes that past demand will predict
future
Many small jobs in cluster so ineffective
Application Aware
Networking
Insight
Application knows every the network
need
So application can in fact instruct the
network
Important Questions
What information do you need?
Size of a flow
Source+destination of the flow
Start time of the flow
Deadline of the flow
Input data
Query HDFS for file size
Proactively
Have the SDN controller tell the job tracker
where to place the end-points
Rack aware placement: reduce inter-rack transfers
Congestion aware placement: reduce loss
Reactive Approach:
FlowComb
Reactive attempt to integrate
bigdata + SDN
No changes to application
Learn information by looking at logs and
determine file size and end-points
Learn information by running agents on
the endhost that determines start times
FlowComb: Architecture
Agents on servers
Hadoop
cluster
Agents
FlowComb
Predictor
Scheduler
Controller
Predictor
Determines size of
intermediate data
Queries Map Via API
Aggregates information
from agents sends to
scheduler
FlowComb: Architecture
Scheduler
Hadoop
cluster
Agents
FlowComb
Predictor
Scheduler
Controller
Open Questions
How about non map-reduce traffic?
Only focus on the active transfers ignores control
msgs and background
CoFlows : Proactive
Approach
Modify the applications
Have them directly inform network of
intent
17
We Want To
Better schedule the network
Intra-coflow
Inter-coflow
18
Coflow APIs
19
Job
finishe
s
terminate(handle)
Coflow
API
Shuffle
finishe
s
Drive
r
put(handle, id, content)
create(SHUFFLE) handle
MapReduc
e
2
0
Summary
Applications know a lot about the transfers
We can reactively learn by using logs
Or modify the application to inform us of these things
CoFlows: proactive
Controls network path, transfer times, and transfer rate
FlowComb: reactive
Controls network paths based on app knowledge
21
ToDo
Need more images from the infobox guys
Maybe improvements and why skethcy
Maybe graphs from flowcomb also