Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation...
-
date post
19-Dec-2015 -
Category
Documents
-
view
220 -
download
0
Transcript of Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation...
![Page 1: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/1.jpg)
Increasing Application Performance In Virtual Environments Through Run-time Inference
and Adaptation
Ananth I. Sundararaj
Ashish Gupta
Peter A. Dinda
Prescience Lab
Department of Computer Science
Northwestern University
http://virtuoso.cs.northwestern.edu
![Page 2: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/2.jpg)
2
Summary• Dynamically adapt unmodified applications on unmodified
operating systems in virtual environments to available resources
• The adaptation mechanisms are application independent and controlled automatically without user or developer help
• Demonstrate the feasibility of adaptation at the level of collection of VMs connected by Virtual Networks
• Show that its benefits can be significant for two classes of applications
![Page 3: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/3.jpg)
3
Outline
• Virtual machine grid computing
• Virtuoso system
• Networking challenges in Virtuoso
• Enter VNET
• VNET, VTTIF Adaptive virtual network
• Evaluation
• Summary
![Page 4: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/4.jpg)
4
Aim
Grid Computing
New Paradigm
Traditional Paradigm
Deliver arbitrary amounts of computational power to perform distributed and parallel computations
Problem1:
Grid Computing using virtual machines
Problem2:
Solution
How to leverage them?
Virtual Machines What are they?
6b
6a
5
4
3b3a
2
1
Resource multiplexing using OS level mechanism
Complexity from resource user’s perspective
Complexity from resource owner’s perspective
Virtual Machine Grid Computing
![Page 5: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/5.jpg)
5
Virtual MachinesVirtual machine monitors (VMMs)
•Raw machine is the abstraction
•VM represented by a single image
•VMware GSX Server
![Page 6: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/6.jpg)
6
The Simplified Virtuoso Model
Orders a raw machine
User
Specific hardware and performance
Basic software installation available
User’s LAN
VM
Virtual networking ties the machine back to user’s home network
Virtuoso continuously monitors and adapts
![Page 7: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/7.jpg)
7
User’s View in Virtuoso Model
User
User’s LAN
VM
![Page 8: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/8.jpg)
8
Outline
• Virtual machine grid computing
• Virtuoso system
• Networking challenges in Virtuoso
• Enter VNET
• VNET, VTTIF Adaptive virtual network
• Evaluation
• Summary
![Page 9: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/9.jpg)
9
User’s friendlyLAN
Foreign hostile LAN
Virtual Machine
VNET: A bridge with long wires
Host
Proxy
X
Virtual NetworksVM traffic going out on foreign LAN
IP network
A machine is suddenly plugged into a foreign network. What happens?
• Does it get an IP address?• Is it a routeable address?• Does firewall let its traffic through? To any port?
![Page 10: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/10.jpg)
10
HostHost
vmnet0
Ethernet Packet Tunneledover TCP/SSL Connection
Ethernet Packet Captured by Interface in Promiscuous mode
“Host Only” Network
Ethernet Packet is Matched against the Forwarding Table on that VNET
Ethernet Packet is Matched against the Forwarding Table on that VNET
First link Second link (to proxy)
Local traffic matrix inferred by VTTIF
Periodically sent to the VNET on the Proxy
VNET
ethz
VM 2
“eth0”
VNET
ethy
IP Network
VM 1“eth0”
vmnet0
A VNET Link
![Page 11: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/11.jpg)
11
Virtual Topology and Traffic Inference Framework (VTTIF) Operation
Application topology is recovered using normalization and pruning algorithms
Ethernet-level traffic monitoring
VNET daemons collectively aggregate a global traffic matrix for all VMs
![Page 12: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/12.jpg)
12
Dynamic Topology Inference by VTTIF
1. Fast updates
Smoothed Traffic Matrix
2. Low Pass Filter Aggregation
3. Threshold change detection
Topology change output
VNET Daemons on Hosts
VNET Daemon at Proxy
AggregatedTraffic Matrix
![Page 13: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/13.jpg)
13
Outline
• Virtual machine grid computing
• Virtuoso system
• Networking challenges in Virtuoso
• Enter VNET
• VNET, VTTIF Adaptive virtual network
• Evaluation
• Summary
![Page 14: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/14.jpg)
14
Monitoring and inference
Application performance measure
Adaptation algorithm
Adaptation mechanisms
AdaptationApplications
Optimization metric
1. Overlay topology
2. Forwarding rules
3. VM migration
1. Single hop
2. Worst fit
1. BSP
2. Transactional ecommerce
1. Application throughput
1. VTTIF
2. Network monitoring
1. Single metric
2. Combined metric
![Page 15: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/15.jpg)
15
Optimization Problem (1/2) Topology Only
Informally stated:
• Input– Network traffic load matrix of application
• Output– Overlay topology connecting hosts– Forwarding rules on the topologySuch that the application throughput is
maximized
The algorithm is described in detail in the paper
![Page 16: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/16.jpg)
16
Foreign hostLAN 1
User’sLAN
Host 2+
VNET
Proxy+
VNET
IP network
Host 3+
VNET
Host 4+
VNET
Host 1+
VNET
Foreign host LAN 3
Foreign host LAN 4
Foreign host LAN 2
VM 1
VM 4VM 3
VM 2
Resilient Star Backbone
Merged matrix as inferred by VTTIF
Illustration of Topology Adaptation in Virtuoso
Fast-path links amongst the VNETs hosting VMs
![Page 17: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/17.jpg)
17
Evaluation• Reaction time of VNET• Patterns: A synthetic BSP benchmark• Benefits of adaptation (performance speedup)
– Eight VMs on a single cluster, all-all topology– Eight VMs spread over WAN, all-all topology
CMU
VM 7University of Chicago
VM 8
Northwestern
VM 1
DOT Network
VM 6
VM 5 …
Wide-Area testbed
Proxy
![Page 18: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/18.jpg)
18
0
0.5
1
1.5
2
2.5
3
3.5
Sec
onds
0.94
1.6
3.23
2.92
2.268
Reaction Time
![Page 19: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/19.jpg)
19
ide
al
com
ple
test
ar 1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Ite
ratio
ns/
seco
nd
Number of Fast Path Links in Virtual Topology
No Fast Path Topology
Full all-to-all network afterstartup measurement+ reconfiguration cost
Full all-to-all frombeginning of run
Dynamic measurement andreconfiguration
Benefits of AdaptationBenefits accrued as a function of the number of fast-path links added
•Patterns has an all-all topology
•Eight VMs are used
•All VMs are hosted on the same cluster
![Page 20: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/20.jpg)
20
ide
al
com
ple
test
ar 1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0
0.5
1
1.5
2
2.5
Ite
ratio
ns/
seco
nd
Number of Fast Path Links in Virtual Topology
No Fast Path Topology
Full all-to-all network afterstartup measurement+ reconfiguration cost
Full all-to-all frombeginning of run
Dynamic measurement andreconfiguration
•Patterns has an all-all topology
•Eight VMs are used
• VMs are spread over WAN
Benefits of AdaptationBenefits accrued as a function of the number of fast-path links added
![Page 21: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/21.jpg)
21
Informally stated:• Input
– Network traffic load matrix of application– Topology of the network
• Output– Mapping of VMs to hosts– Overlay topology connecting hosts– Forwarding rules on the topologySuch that the application throughput is maximized
Optimization Problem (2/2) Topology + Migration
The algorithm is described in detail in the paper
![Page 22: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/22.jpg)
22
Evaluation
• Applications– Patterns: A synthetic BSP benchmark– TPC-W: Transactional web ecommerce
benchmark
• Benefits of adaptation (performance speedup)– Adapting to compute/communicate ratio– Adapting to external load imbalance
![Page 23: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/23.jpg)
23
Effect on BSP Application Throughput of Adapting to Compute/Communicate Ratio
![Page 24: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/24.jpg)
24
Effect on BSP Application Throughput of Adapting to External Load Imbalance
![Page 25: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/25.jpg)
25
TPCW Throughput (WIPS) With Image Server Facing External Load
No Topology Topology
No Migration 1.216 1.76
Migration 1.4 2.52
![Page 26: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/26.jpg)
26
Outline
• Virtual machine grid computing
• Virtuoso system
• Networking challenges in Virtuoso
• Enter VNET
• VNET, VTTIF Adaptive virtual network
• Evaluation
• Summary
![Page 27: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/27.jpg)
27
Summary• Dynamically adapt unmodified applications on unmodified
operating systems in virtual environments to available resources
• The adaptation mechanisms are application independent and controlled automatically without user or developer help
• Demonstrate the feasibility of adaptation at the level of collection of VMs connected by Virtual Networks
• Show that its benefits can be significant for two classes of applications
![Page 28: Increasing Application Performance In Virtual Environments Through Run-time Inference and Adaptation Ananth I. Sundararaj Ashish Gupta Peter A. Dinda Prescience.](https://reader036.fdocuments.us/reader036/viewer/2022062714/56649d3b5503460f94a15e2b/html5/thumbnails/28.jpg)
28
• Future Work– Free network measurement (Wren) – Collaboration with CS, W&M– Applicability of a single optimization scheme
• Related Talk at HPDC 2005– J. Lange, A. Sundararaj, P. Dinda, “Automatic Dynamic Run-time
Optical Network Reservations”– Wednesday, July 27, 2:00 P.M.
• Please visit– Prescience Lab (Northwestern University)
• http://plab.cs.northwestern.edu
– Virtuoso: Resource Management and Prediction for Distributed Computing using Virtual Machines
• http://virtuoso.cs.northwestern.edu• VNET is publicly available from above URL
For More Information