Free%20Network%20Measurement%20for%20Adaptive%20Virtualized%20Distributed%20Computing

1
Free Network Measurement for Adaptive Virtualized
Distributed Computing

• Ashish Gupta, Marcia Zangrilli, Ananth
  Sundararaj, Anne Huang, Peter A. Dinda, Bruce B.
  Lowekamp

2
Overview
Benefits of VMs transparent portability,
adaptation, security
Virtual Machines

• Contributions
• Online passive measurement of physical layers
  available bandwidth (Wren)
• Integration of Virtuosos application monitoring
  and Wrens traffic monitoring
• Adaptation algorithms that use passive monitoring
  to solve challenging adaptation problems

Virtual Network
Physical Network
3
Adaptive Virtualized Distributed Computing

• How can we efficiently utilize resources in a
  virtual machine distributed system?
• Accurately monitor resource availability
• Transparently adapt to changing conditions
• Keep application portability simple

4
Claim

• Virtualization enables the broad application of
  dream techniques
• Adaptation
• Resource reservation
• using existing, unmodified applications and
  operating systems
• So everyone can use the techniques

5
Optimization of Virtual System Environment
Benefit Completely independent of application or
Operating System
6
Outline

• Virtuoso
• Overview of distributed VM system
• VTTIF
• VNET
• Wren
• Online Wren overview
• Wren performance
• Integration of Virtuoso and Wren
• Adaptation
• Algorithms
• Results

7
Virtuoso
Distributed computing environment composed of
virtual machines interconnected with virtual
networks

• Automatically infer application demands
  (network/CPU)
• Monitor resource availability (bw/latency/CPU)
• Adapt distributed application for better
  performance/cost effectiveness
• Reserve Resources when possible

8
Application communication topology and traffic
load application processor load
VM Layer
Vnetd layer can collect all this information as
a sideeffect of packet transfers
Vnetd Layer
and invisibly act

• VM Migration

• Topology change

• Routing change

• Reservation

Network bandwidth and latency sometimes topology
Physical Layer
9
Virtual Topology and Traffic Inference Framework
(VTTIF) Operation

• Infers application topology and traffic load at
  runtime
• Resistant to rapid fluctuations and provides
  damped network view
• All local views aggregated to central proxy to
  give global view of distributed application

10
Virtual Topology and Traffic Inference Framework
(VTTIF) Operation
Ethernet-level traffic monitoring
VNET daemons collectively aggregate a global
traffic matrix for all VMs

• Application topology is recovered using
• normalization and pruning algorithms

11
VNET

• Virtual overlay network ? creates illusion of LAN
  over wide area
• Network transparency with VM migration
• Ideal monitoring point for application monitoring

12
Watching Resources from the Edge of the Network
(Wren) A Hybrid Monitoring Approach

• Wren Design
• Kernel-level instrumentation to collect traces of
  application traffic.
• Analysis and management of traces handled in
  user-level.
• Wren capabilities
• Observes incoming/outgoing packets
• Online analysis to derive latency/bandwidth
  information for all host pair connections
• Answers network queries for any pair of hosts

13
Wren Architecture
14
Wren Online Available Bandwidth Algorithm

• Applies self-induced congestion principle
• If packets are sent at a rate larger than the
  available bandwidth, the queuing delays will have
  an increasing trend.
• Find the rate just before queuing delays are
  incurred
• Identifies outgoing Maximal length trains with
  similar spaced packets.
• Calculates ISR ( Initial Sending Rate ) for these
  trains.
• Monitors ACK return rate to determine trends in
  RTTs.
• Increase trend indicates congestion, non
  increasing trend indicates lower bound for bw.

15
Wren Performance
Controlled load/latency testbed Nistnet ? emulate
WAN environment with congestion Latency 20 to
100 ms , bw 3 to 25 Mbps
Key Advantage WREN accurately reports
available bandwidth when application traffic does
not saturate the path
16
Integrating Virtuoso and Wren
Application
Guest OS Kernel
Virtual Machine
Virtual Machine Monitor
VADAPT Adaptation
VTTIF Application Inference
Wren Network Inference
TCP / UDP Forwarding
Layer 2 Network Interface
Host OS Kernel
LAN
Other VNET daemon
17
Adaptation Process
18
What defines Good Adaptation?

• Various ways to define good adaptation

Current Metric Maximum residual bottleneck
bandwidth How can we map the processes and paths
such that (available bandwidth demanded
bandwidth) is maximized ? ? Maximum room for
performance improvement
19
Optimization Problem

• Given the
• network traffic load matrix of the application
• computational intensity in each VM
• topology of the network
• load on its links, routers and hosts
• What is the
• mapping of VMs to hosts
• overlay topology connecting the hosts
• forwarding rules on that topology
• required CPU and network reservations
• That
• maximizes the application performance?

20
Problem formulation
Measured data
Application demands
Constraints
Objective function
21
Greedy Heuristic

• Mapping
• Identifies Hosts which have good bandwidth
  connectivity and maps VMs over them
• Overlay paths
• Uses adapted Dijktra to find widest paths
  depending on bandwidth demands of application
  process pairs (sorted in decreasing order)
• ? finds path which leaves maximum residual
  bottleneck bandwidth

22
Simulated Annealing

• Motivation Search Space is very large ? Huge
  number of possibilities for mapping and overlay
  paths
• Approach
• Start with an initial solution
• Perturb current configuration and evaluate with a
  cost function
• Continue Controlled Perturbation until a good
  cost function is achieved

Perturbation function and algorithm details in
paper
23
Experimental Setup

• Evaluation conducted in simulation
• In each scenario the goal is
• to generate a configuration consisting of VM to
  Host mappings
• paths between the communicating VMs
• Such that the total residual bottleneck bandwidth
  is maximized
• We compare
• greedy heuristic (GH)
• simulated annealing approach (SA)
• SA with the GH solution as the starting point
  (SAGH).
• Additionally we also maintain the best solution
  found so far with (SAGH), i.e. (SAGHB), where
  B indicates the best solution so far.

24
Adaptation Results
Scenario 1 Only a particular VM to Host mapping
yields good performance.
25
Scenario 1 Results

• Both Annealing and Greedy perform well.
• Annealing advantage Multi-Constraint
  optimization easy

26
Scenario 2 Large 256 host topology. 32
potential hosts, 8 Virtual Machines

• Results for Multi Constraint Cost Function
  Bandwidth and Latency
• Annealing easy to adapt and finds good mappings
  compared to heuristic

27
Conclusion

• Network measurements can be provided for free!
• These measurements can be used to improve
  application performance through adaptation
• Virtuoso and Wren Integrated system
• Low overhead
• Provides application and resource measurements
• Allows transparent optimization of application
  performance
• Adaptation Strategies
• Greedy heuristic and simulated annealing
  approaches are able to find good
  mappings/configurations

28
For More Information

• Please visit
• Prescience Lab (Northwestern University)
• http//plab.cs.northwestern.edu
• Wren Watching Resources fro the Edge of the
  Network (William and Mary)
• http//www.cs.wm.edu/lowekamp/wren.html
• Virtuoso Resource Management and Prediction for
  Distributed Computing using Virtual Machines
• http//virtuoso.cs.northwestern.edu
• VNET is publicly available from above URL