November 26, 2002 TeleSim Research Group 1
Performance and Robustness Testingof Wireless Web Servers
Guangwei BaiKehinde OladosuCarey Williamson
November 26, 2002 TeleSim Research Group 2
1. Introduction and Motivation Observation: the same wireless technology that allows a Web client to be mobile also allows Web servers to be mobile Idea: portable, short-lived, ad hoc networks Possible applications:
o classroom area networks, seminarso press conferences, media eventso sporting events, gaming, exhibitionso conferences and trade showso disaster recovery sites, field work, etc.
November 26, 2002 TeleSim Research Group 3
Background: Portable Networks Assumptions: the characteristics of a portable short-lived network are:
o set it up when needed; tear down aftero only needed for minutes or hourso when may not be known a priorio where may not be known a priorio no existing infrastructure of any kindo general Internet access not availableo general Internet access not requiredo pre-defined content; target audienceo 1-100 users; mobile; limited bw needed
November 26, 2002 TeleSim Research Group 4
2. Objectives to assess feasibility of portable networks to benchmark the performance capabilities and limitations of an Apache Web server in a wireless ad hoc network to identify the performance bottlenecks to understand impacts of different factors
o number of clientso Web object sizeo persistent connectionso transmit power (energy consumption)o wireless channel conditions
November 26, 2002 TeleSim Research Group 5
3. Experimental Setup
• Compaq Notebooks (1.2GHz Pentium III, 128MB RAM, 512 KB L2 cache, Cisco Aironet 350 network cards)• RedHat Linux 7.3, httperf, Apache 1.3.23, SnifferPro 4.6• Network: 11 Mbps IEEE 802.11b wireless LAN, ad hoc mode
November 26, 2002 TeleSim Research Group 6
Experimental Setup (Cont’d)
• IEEE 802.11b: a standard for wireless LANs Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA), up to 11 Mbps data rate at physical layer
• ad hoc modeframes are addressed directly from sender to receiver
• httperf Web benchmarking software tool developed at HP Labs
• Web server: Apache (version 1.3.23)Process-based, flexible, powerful, HTTP/1.1-compliant
• SnifferPro 4.6real-time capture, recording all wireless channel activity,enabling protocol analysis at MAC, IP, TCP and HTTP layers
7
4. Experimental Design
Factor Levels Number of Clients 1, 2, 4HTTP Transaction Rate (per-client) 10, 20, 30, …, 160
HTTP Transfer Size (KB) 1, 2, 4, 8, …, 100Persistent Connections no, yesHTTP Requests per Connection 1, 5, 10, 15, …, 60Transmit Power (mW) 1, 5, 20, 30, 50, 100
Client-Server Distance (m) 1, 10, 100
• Impacts of different factors on wireless Web server performance (one-factor-at-a-time)
Experimental Factors and Levels
• Performance metrics– HTTP transaction rate, throughput, response time, error rate at Application Layer,– TCP connection duration at Network Layer– Transmit queue behaviour at Link Layer,
November 26, 2002 TeleSim Research Group 8
5. Measurement Results and Analyses - Expt 1: Request Rate - Expt 2: Transfer Size - Expt 3: Number of Clients - Expt 4: Persistent Connections - Expt 5: Transmit Power - Expt 6: Wireless Channel
November 26, 2002 TeleSim Research Group 9
Experiment 1: Request Rate
Purpose: to determine the range of feasible and sustainable loads for the wireless Web server
Design: • Number of Clients: 1• HTTP transaction rate: 10, 20, …, 160 req/sec• HTTP transfer size: 1 KB (fixed)• Persistent connections: no• Transmit power: 100 mW• Client-server distance: 1 meter (on same desk)
November 26, 2002 TeleSim Research Group 10
Wireless Web Performance at Application Layer
Main observation:• As the offered load increases: linear increase instability lower plateau• Peak throughput < 1 Mbps for 1 KB transfers
November 26, 2002 TeleSim Research Group 11
Transmit Queue Behaviour for Experiment 1
Main observation: Wireless LAN is the bottleneck• Packet drops occur from link-layer queue (client side)• Even before they get on the wireless LAN!!!Reason:• No flow control / backpressure mechanism• Note: default queue size is 100 in the Linux kernel
November 26, 2002 TeleSim Research Group 12
Wireless Web Performance at Application Layer (Cont’d)
Main observation:• the response time is about 9 ms at low load, increase significantly to over 2 sec at high load (>85 req/sec)• failures occur frequently under overload
November 26, 2002 TeleSim Research Group 13
Measurement at Network Layer
Overload: 100 req/secQueue buildup,Packet drops, Retransmissions,TCP resets
Low load: 10 req/secStable performance Mean: 9.7ms
Medium load: 50 req/secGreater variation, 2 spikes Mean: 10ms
High load: 80 req/secMore variability, some spikes, slight skew
November 26, 2002 TeleSim Research Group 14
Experiment 2: Transfer Size
Purpose: to study impact of HTTP response size
Design: • Number of Clients: 1• HTTP transaction rate: 10 req/sec (fixed)• HTTP transfer size (KB): 1, 2, 4, 8, …• Persistent connections: no• Transmit power: 100 mW• Client-server distance: 1 meter (on same desk)
November 26, 2002 TeleSim Research Group 15
Measurement at Network Layer
General observation:
as HTTP transfer sizeincreases, mean TCP connection duration increases, as does thevariance of distribution.
November 26, 2002 TeleSim Research Group 16
Light load: 8 KBDuration: 24 msec Throughput: 2.8 Mbps
Overload: 64 KBDuration: >100 msec Throughput: 4.1 Mbps
Medium load: 32 KBDuration: 67 msec Throughput: 3.9 Mbps
Measurement at Network Layer
November 26, 2002 TeleSim Research Group 17
Experiment 3: Number of Clients
Purpose: to study impact of high load generated by multiple clients
Design: • Number of Clients: 2, 3, 4• HTTP transaction rate: 10, 20, …, 160 req/sec• HTTP transfer size: 1 KB (fixed)• Persistent connections: no• Transmit power: 100 mW• Client-server distance: 1 meter (on same desk)
18
Wireless Web Performance at Application Layer (4 Clients)
November 26, 2002 TeleSim Research Group 19
Main observation:• 4 clients share network and server resources equally• 30% higher aggregate throughput (110 conns/sec)• bottleneck is now at server network card (drops!!)
Wireless Web Performance at Application Layer (4 Clients)
November 26, 2002 TeleSim Research Group 20
Wireless Web Performance at Application Layer (2 or 3 Clients)
November 26, 2002 TeleSim Research Group 21
Wireless Web Performance at Application Layer (2 or 3 Clients)
Main observation: unfairness problem at high loads:
one client obtained a higher proportion of the throughput at expense of another (don’t know why?)
November 26, 2002 TeleSim Research Group 22
Experiment 4: Persistent Connections
Persistent Connections:• Multiple HTTP transactions can be sent on the same TCP connection.• amortize overhead of TCP connection processing• reduce memory consumption for TCP state
Purpose of this experiment: to study impact of persistent connection on wireless Web performanceDesign:
• Number of Clients: 1 and 2• HTTP transaction rate: 10 req/sec (fixed)• HTTP transfer size: 1 KB (fixed)• Persistent connections: yes• Transmit power: 100 mW• Client-server distance: 1 meter (on same desk)
November 26, 2002 TeleSim Research Group 23
Achieved Throughput for Experiment with Persistent Connections
Main observation:• Peak throughput: 3.22 Mbps, 3.5x improvement over non-persistent connections (0.9 Mbps),• two clients share the server and network resources equally
November 26, 2002 TeleSim Research Group 24
Experiment 5: Transmit Power
Energy consumption- an important issue for mobileClients and Server.
Purpose: to see what transmit power is required for acceptable performance in classroom setting
Design: • Number of Clients: 1• HTTP transaction rate: 10 req/sec (fixed)• HTTP transfer size: 1 KB (fixed)• Persistent connections: no• Transmit power: 1, 5, 20, 100 mW• Client-server distance: 10 meter (same floor)
November 26, 2002 TeleSim Research Group 25
Measurement at Network Layer
General observation:
If transmit power<10 mW:• MAC-layer retransmits• rightward skew• unacceptable perf.
If transmit power20 mW:• acceptable performance
November 26, 2002 TeleSim Research Group 26
Experiment 6: Wireless Channel Characteristics
Wireless Internet is characterized by limitedbandwidth, high error rates, and interference.
Purpose: to study the impact of the wireless channel characteristics on wireless Web performance
Design: • Number of Clients: 1• HTTP transaction rate: 10 req/sec (fixed)• HTTP transfer size: 1 KB (fixed)• Persistent connection: no• Transmit power: 100 mW• Client-server distance: 1m, 10m, 100m
November 26, 2002 TeleSim Research Group 27
Low load: 10 req/secSignificant skew to thetail of the distribution,Some periodicity (why?)
Medium load: 50 req/secSignificant skew to thetail of the distribution
Measurement at Network Layer (100m scenario)
November 26, 2002 TeleSim Research Group 28
6. Summary and Conclusions
What we did: wireless Web server, portable nw• Application-layer measurements (httperf)• Network-layer measurements (Wireless Sniffer)
Our results show:• Server capability: 100 conn/sec for non-persistent HTTP with throughputs up to 4 Mbps (adequate?)• Bottleneck: at wireless network interface• Some “network thrashing” for large HTTP transfers when the network utilization is high (aborts, resets)• Effect of wireless channel on performance at TCP and HTTP-level (MAC-layer retransmits)• Power consumption issue for mobile client and server
November 26, 2002 TeleSim Research Group 29
7. Future Work
Explaining the anomalies (fairness, periodicity) Better system instrumentation (Linux) More realistic Web workloads Larger WLAN testing (classroom scenario) Repeat experiments with IEEE 802.11a (55 Mbps) Kenny’s M.Sc. Thesis... Another paper?