Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project...

22
Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001

Transcript of Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project...

Page 1: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Maintaining Performance while Saving Energy on Wireless LANs

Ronny Krashinsky

6.929 Term Project

12-7-2001

Page 2: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Motivation

• Mobile devices limited by battery weight and lifetime• Wireless network access consumes a lot of energy• Want to disable the network interface card whenever its

not in use– Basic problem: data may arrive from the network at any time

• Focus of this work: a mobile client communicating with a wired base-station to perform request/response traffic (e.g. web browsing)– Not focusing on: ad hoc networks, mobile servers, real-time

communication (voice)

– Not relying on high-level knowledge of application state

Page 3: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

802.11 Power-Saving Mode Overview(For Infrastructure Networks)

• Network Interface Card power consumption:– Cisco Aironet: 1.7W Tx, 1.2W Rx, 1.1W Idle, 50mW Sleep

• Basic idea: sleep to save energy, periodically wakeup to check for pending data

• Clients go to sleep after sending or receiving data• Base-station buffers received data while client is asleep• Base-station sends out beacons every 100ms indicating

whether or not the Client has pending data• Client wakes up to listen to beacon, then polls Base-station to

receive data (ListenInterval can be less than BeaconPeriod)• Client can wake up to send data at any time

Page 4: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Talk Outline

• Measured performance of TCP over 802.11 PSM (it’s not good)

• Trace analysis for characteristics of client HTTP traffic (how to save energy)

• Proposed enhancements to 802.11 PSM to improve performance and minimize energy

• Simulation of web browsing traffic over existing 802.11 PSM and alternatives

Page 5: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Request/Response Over TCP Over 802.11

RTT+delta

MobileClient

Base-Station Server

syn

syn+ack

ackrequest

responsestart

RTT

RTT

RTT

PSM Off MobileClient

Base-Station Server

syn

syn+ack

ackrequest

responsestart

100ms

100ms

100ms

sleep

sleep

sleep

beacon

PSM On

queue

beacon

beacon

queue

queue

Page 6: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Request/Response Performance Test

for (N := various sizes) {

start timer

for (several iterations) {

TCP connect to server

send request

receive N bytes

close connection

}

stop timer

}

•Client:Compaq iPAQ with Enterasys Networks RoamAbout 802.11 NIC

•Servers:

•Methodology:repeat tests five times, alternating between PSM on and off, use mean

RTT Bandwidth

LCS 5ms 10Mbps

Berkeley 80ms 10Mbps

Home (DSL)

50ms 70Kbps

Page 7: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

802.11 PSM Measured Performance

Page 8: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

802.11 PSM Measured Slowdown

Conclusion: 802.11 PSM is too coarse-grain to maintain network performance

Page 9: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Client Network Usage

responseresp

responseresponse

wait recv idle wait receive idle

requestReq/Resp 1:Req/Resp 2:Req/Resp 3:Req/Resp 4:

time

Client State:

• Analyzed UC Berkeley Home-IP (modem) HTTP Traces:client ID, request time, response start time, response end time

• Classified client state as: {wait, idle, receive}• Discarded incomplete transactions (no timestamp)• Ignored receive and idle times longer than 1000s

Page 10: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Client Network Usage CharacteristicsWait Time Idle Time

Conclusion: 802.11 PSM is too fine-grain to reduce energy effectively

• Most wait time and idle time is spent in a few number of long latency events

• These events will therefore account for most of the sleep energy

Page 11: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Proposed Solution: StayAlive and ListenInterval-Backoff

0s 1s 2s 3s

requestPSM basic

wakeup to listen to beacons…

StayAlive

stay awake after sending request

Listen-IntervalBackoff

increase ListenInterval if there is no response

max = 0.9s

20% delay

never sleep for more than 20% of total time since request

Page 12: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Latency and Energy Comparison

Latency (vs. No PSM)

Energy (vs. PSM basic)

Short Medium Long active (awake)

listening to beacons

PSM basic Increased by up to 100ms

StayAlive Unchanged Increased by up to 100ms Increased Unchanged

ListenInterval-Backoff (2x)

Increased by up to 2x Increased by up to 0.9s

Unchanged Decreased

20% delay Unchanged Increased by up to

20%

Increased by up to 0.9s

Increased Decreased

Page 13: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Client Web Browsing Simulation• Modeled 802.11 PSM using ns-2

– Did not model detailed MAC protocol: no channel contention, no node movement, no packet losses

– Modified Link C++ code to support sleep mode and send alerts to OTcl , control and beaconing in OTcl

• Modeled HTTP traffic using empirical model– Based on study by Bruce Mah

– Limited “Think Time” to 1000s

– Added “Server Response Time” based on wait time from UCB Home-IP traces (less 100ms to account for network delays).

– Updated to use FullTcp

• Client BaseStation: 0.1ms, 5Mbps

• BaseStation Server: 20ms, 10Mbps

• Energy: 1W while active, 50mW while sleeping, 5mJ per listened-to beacon (1W5ms)

Page 14: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Performance Results

Page 15: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Performance and Energy Results

energy per page(PSM off = 54 J)

PSM basic StayAlive LI-Backoff: 2x Max %delay

slowdown(vs. PSM off)

Page 16: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Conclusions• Existing 802.11 PSM causes RTTs to be rounded up to

the nearest 100ms– This adversely affects short TCP connections which are

limited by the RTT

– A viable solution is to stay awake for a short period of time after sending a request

• When using 802.11 PSM, almost all energy consumption is due to sleep power and listening to beacons– ListenInterval-Backoff can reduce the listen energy

– Longer sleep intervals have the potential to enable deeper sleep modes

Page 17: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.
Page 18: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

(backup slides)

Page 19: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Simulation vs. Measured

Page 20: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Actual Values Used in HTTP Simulation

Page 21: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.
Page 22: Maintaining Performance while Saving Energy on Wireless LANs Ronny Krashinsky 6.929 Term Project 12-7-2001.

Intersil PRISM Radio Chip Set

Current (mA)

WakeupTime (s)

Tx 488

Rx 287

PSM 1 190 1

PSM 2 70 25

PSM 3 60 2000

PSM 4 30 5000