+ All Categories
Home > Documents > The Effects of Wide-Area Conditions on WWW Server Performance

The Effects of Wide-Area Conditions on WWW Server Performance

Date post: 31-Jan-2016
Category:
Upload: trang
View: 22 times
Download: 0 times
Share this document with a friend
Description:
The Effects of Wide-Area Conditions on WWW Server Performance. Erich Nahum, Marcel Rosu, Srini Seshan, Jussara Almeida. IBM T.J. Watson Research Center, CMU, Univ. of Wisconsin. switch. server. clients. Problem: real Internet doesn’t work this way!. Motivation: Benchmarking Today. - PowerPoint PPT Presentation
Popular Tags:
30
The Effects of Wide-Area Conditions on WWW Server Performance Erich Nahum, Marcel Rosu, Srini Seshan, Jussara Almeida IBM T.J. Watson Research Center, CMU, Univ. of Wisconsin
Transcript
Page 1: The Effects of Wide-Area Conditions on WWW Server Performance

The Effects of Wide-Area Conditions on WWW Server

Performance

Erich Nahum, Marcel Rosu, Srini Seshan, Jussara Almeida

IBM T.J. Watson Research Center, CMU, Univ. of Wisconsin

Page 2: The Effects of Wide-Area Conditions on WWW Server Performance

Motivation: Benchmarking Today

clients

switch

server

Problem: real Internet doesn’t work this way!

Page 3: The Effects of Wide-Area Conditions on WWW Server Performance

Motivation: Real Life

clients

server

Internet

Evaluate WWW server performance under WAN conditions

Page 4: The Effects of Wide-Area Conditions on WWW Server Performance

Web Server Performance

• Workload Generators Webstone, SpecWeb, SURGE, s-client, httperf, etc.+ Based on measured traffic behavior+ Reproducible- WAN case is ignored: no drops, delays, etc.

Want an environment that is *both* realistic *and* reproducible

• Live Server Analysis California elections, 96 Olympics, WAWM+ Capture real WAN conditions- But not reproducible

Page 5: The Effects of Wide-Area Conditions on WWW Server Performance

Outline

• Motivation and Background• The WASP Environment

– Hardware and software– Workload generators

• Results– Effects of packet loss– Effects of packet delay– Effects of TCP variants

• Summary and Conclusions

Page 6: The Effects of Wide-Area Conditions on WWW Server Performance

Wide-Area Server Performance

• What WASP is not:– Doesn’t reproduce a specific web site– Doesn’t reproduce a specific network topology

• What WASP is:– Realistic: emulates the WAN environment– Reproducible: allows iterative analysis– Configurable: can vary many parameters– Scalable: scales to very large workloads

A testbed for server performance analysis

Page 7: The Effects of Wide-Area Conditions on WWW Server Performance

Centralized Approach

GigabitEthernetswitch

server

clients

WAN emulator used to drop & delay packets

WANemulator

100 Base-T 1 Gbps

Page 8: The Effects of Wide-Area Conditions on WWW Server Performance

WASP Approach

• Each client acts as a ‘WAN emulator’• Use DummyNet to drop and delay packets

User AppSocket

TCP

IPDummyNet

Ethernetclientdelay

drop

Page 9: The Effects of Wide-Area Conditions on WWW Server Performance

Scaling with Load

Centralized approach doesn’t scale

Page 10: The Effects of Wide-Area Conditions on WWW Server Performance

Packet Loss Model

• Two-state loss model based on work by Bolot 93, Paxson 97, Rubenstein et al. 2000

• Packets forwarded in good state, dropped in bad• Transitions based on desired loss rate

Good Bad

loss event probability

(1 - loss event probability)

conditional lossprobability

(1 - conditional loss probability)

Page 11: The Effects of Wide-Area Conditions on WWW Server Performance

Workload Generators

• S-client (from Rice), SURGE (from BU) • WaspClient integrates the two

Responses Requests

Page 12: The Effects of Wide-Area Conditions on WWW Server Performance

Putting it all together

clients

switch

server

Web server software( Apache, Flash)

200 MHz PowerPC w/AIX 4.3.3

Workload generator

(WaspClient)

500 MHz P/3w/ FreeBSD

3.3 & DummyNet

GigabitEthernet

FastEthernet

Page 13: The Effects of Wide-Area Conditions on WWW Server Performance

Experimental Methodology

• Performance Metrics:– Server throughput, utilization, response time,

capacity

• Sensitivity Analysis:– Vary generated load in SURGE UE’s– Vary loss rate from 0 to 9 %– Vary RTT from 0 to 400 msec– Parameters taken from Paxson97, Allman2000

• Methodology:– Average of 10 runs– Each run lasts 10 minutes– 90 % confidence intervals

Page 14: The Effects of Wide-Area Conditions on WWW Server Performance

Outline

• Motivation and Background• The WASP Environment

– Hardware and software– Workload generators

• Results– Effects of packet loss– Effects of packet delay– Effects of TCP variants

• Summary and Conclusions

Page 15: The Effects of Wide-Area Conditions on WWW Server Performance

Throughput vs. Loss Rate

Throughputs fall with increasing loss

Page 16: The Effects of Wide-Area Conditions on WWW Server Performance

Utilization vs. Loss Rate

Utilization falls with increasing loss

Page 17: The Effects of Wide-Area Conditions on WWW Server Performance

What’s going on?

pR

BT

*

*3/25.1

Simple model for TCP throughput, where:B = max segment size (MSS), R = round-trip time, andp = loss rate.

More elaborate models available from:Padhye et al. (SigComm98), Cardwell et al. (Infocom2000)

Page 18: The Effects of Wide-Area Conditions on WWW Server Performance

Latency vs. Loss Rate

Latency increases with loss rate

Page 19: The Effects of Wide-Area Conditions on WWW Server Performance

Capacity vs. Loss Rate

Capacity falls with loss rate

Page 20: The Effects of Wide-Area Conditions on WWW Server Performance

Outline

• Motivation and Background• The WASP Environment

– Hardware and software– Workload generators

• Results– Effects of packet loss– Effects of packet delay– Effects of TCP variants

• Summary and Conclusions

Page 21: The Effects of Wide-Area Conditions on WWW Server Performance

Throughput vs. RTT

Throughputs decrease with RTT

Page 22: The Effects of Wide-Area Conditions on WWW Server Performance

Utilization vs. RTT

Utilization falls with increasing RTT

Page 23: The Effects of Wide-Area Conditions on WWW Server Performance

Latency vs. RTT

Latency increases with larger RTT’s

Page 24: The Effects of Wide-Area Conditions on WWW Server Performance

Capacity vs. RTT

Capacity falls slightly with RTT

Page 25: The Effects of Wide-Area Conditions on WWW Server Performance

Many Variants of TCP:

• Reno (current baseline in the Internet):– Coarse-grained timeouts, fast retransmit– Recovers 1 lost segment every 3 RTT’s

• New Reno:– Uses partial acknowledgement to improve loss recovery– Recovers 1 lost segment every RTT– Sender-side only modification

• Selective Acknowledgements (SACK):– Uses SACK option bit field to improve loss recovery– Recovers up to 3 segments per RTT– Requires modifications to both sender and receiver

• Other schemes exist (e.g., Vegas)

How do variants affect server performance?

Page 26: The Effects of Wide-Area Conditions on WWW Server Performance

TCP Variants: Latency

SACK provides lower latency

Page 27: The Effects of Wide-Area Conditions on WWW Server Performance

Summary• Presented the WASP environment

– Emulates WAN conditions in a controlled setting– Scalable, reproducible, configurable

• Several results:– Delays and losses affect performance– Loss reduces capacity, increases latency– Delays increase latency but not capacity– SACK, New Reno can reduce response time, don’t affect capacity

• Other fallout:– Used to find bugs in AIX, Flash, AFPA (IBM server)– Convinced AIX group to deploy SACK & New Reno

Benchmarks must include WAN characteristics

Page 28: The Effects of Wide-Area Conditions on WWW Server Performance

Future Directions

• HTTP 1.1• Linux• Bandwidth limitations• Dynamic content• Other workloads:

– Proxies– Clients– SSL

WASP provides a general environment for performing all kinds of evaluations

Page 29: The Effects of Wide-Area Conditions on WWW Server Performance

Apache Capacity vs. Loss

Capacity decreases with loss rate

Page 30: The Effects of Wide-Area Conditions on WWW Server Performance

Apache Capacity vs. RTT

RTT doesn’t really affect capacity


Recommended