Upload
barnard-webster
View
217
Download
0
Embed Size (px)
Citation preview
SwitchR: Reducing System Power Consumption
in a Multi-Client Multi-Radio Environment
Yuvraj Agarwal (University of California, San Diego)
Trevor Pering, Roy Want (Intel Research),
Rajesh Gupta (UC San Diego)
Wearable and Mobile Devices:
• Increasing Functionality – Faster processors, more memory
• Applications are increasingly communication intensive– Streaming video, VoIP, Downloading files
• Multiple wireless radios often integrated on single device – (Bluetooth for PANs, WiFi for high-bandwidth data access)
• Wearable/Mobile Computers Power Consumption is very important! – Limited by battery lifetime
– Communication over WiFi reduces battery lifetime even further…. In some cases up to 50% of total energy drain!
2
Reducing the energy for communication
• Opportunity: Availability of multiple radio interfaces …– Can all be used for data transfer
– Different characteristics : bandwidth, range, power consumption
• Typically function as isolated systems,
– Can we coordinate usage to provide a unified network connection ?
Seamlessly switch between radios
– Primary Goal: Save energy
3
+ X
Radio Characteristics
4
050
100150200250300350400450
Zigbee BT 802.11
Idle
Po
we
r (m
W)
0
50
100
150
200
250
En
erg
y/B
it (
nJ
/bit
)
0.25Mbps 1.1Mbps 11Mbps
Higher throughput radios have a lower energy/bit value … have a higher idle power consumption …and they have different range characteristics
Multi-Radio Switching
• CoolSpots [Mobisys ‘06]: – Multi-Radio switching for a single-client scenario
– Specialized access point (Bluetooth + WiFi)
– Switching decisions – Local to client
• SwitchR: – Leverage existing WiFi APs : Incrementally deployable
– Considers traffic imposed by other devices in a multi-client scenario
– Switching decision – global since it affect other clients
– Evaluate energy savings on a distributed testbed
5
Problem Statement: Reduce energy consumption by choosing appropriate radio interface, while taking into consideration other clients.
SwitchR Architecture
6
Bluetooth Link
WiFi Link
Ethernet Link
Wi-Fi AP (WFAP)
Infrastructure Network
Wi-FiZone
MD1
MD3
MD4
MD2
BTG (Bluetooth Gateway)
MD = Mobile Devices
Switching Mechanism: • Network Level Reconfigurations• ARPs and Routing updates
Switching Policy: • Hybrid Approach • Application requirements at nodes (local) • Channel quality and bandwidth (global)
Multi-Client Switching Policy
• Hybrid approach to make switching decisions – Local knowledge (node level)
– Global (channel utilization by other nodes)
• Switching up (Bluetooth WiFi) – ICMP response time and radio RSSI values
– Capture application needs and channel characteristics
• Switching-down (WiFi Bluetooth) – Measure application bandwidth requirements
– Periodically query BTG for residual capacity
– Measure channel/link quality (local)
7
Evaluation: Testbed
8
Bluetooth (Always Connected)
WiFi (Dynamically Switched)
Static Wired Connection
Wi-Fi AP
Infrastructure Network
Wi-FiZone
MD1
MD3MD4
MD2
BTG (Bluetooth Gateway)
Mobile Device (MD)
• Stargate2 research platform• WiFi + Bluetooth + Integrating power and data monitoring
• Benchmark applications are striped across devices
Stargate2 node
Evaluation: Benchmarks
9
Baselines:• Idle: connected, but no data transfer• Transfer: bulk TCP data transfer
Web: • Combination of idle and data transfer
• Idle: “think time”
• Small transfer: basic web-pages
• Bulk transfer: documents or media
Streaming:
• Media: 128k, 156k and g711 VoIP codec
• Various QoS requirements
Evaluation: Switching Policies
• Baselines policies– “Wifi-CAM” (Awake Mode)
– “Wifi-PSM” (Power Save Mode)
• Single-Client based “cap-dynamic” switching policy
• SwitchR: “multi-client” switching policy – Combines both local (per client) and global knowledge
10
Results: Baselines
11
Basic Benchmark Suite
0
0.2
0.4
0.6
0.8
1
1.2
Ave
rage
Pow
er (
Nor
mal
ized
)
00.20.40.60.811.21.41.61.82
Ene
rgy-
per-
bit
(Nor
mal
ized
)Bluetooth-Power WiFi-power Energy-Per-Bit
g711 web transferidle
Switching policies perform better that WiFi policies for “idle” benchmark, similar for “transfer”
Results:
12
Loaded Benchmark Suite
0
0.2
0.4
0.6
0.8
1
1.2
Ave
rage
Pow
er (
Nor
mal
ized
)
00.20.40.60.811.21.41.61.82
Ene
rgy-
per-
bit
(Nor
mal
ized
)Bluetooth-Power WiFi-power Energy-Per-Bit
g711 stream128 kbps
stream156 kbps
web
multi-client policy saves up to 62% over single-client cap-dynamic policy VoIP and streaming benchmarks benefit most since streams can use BT channel
Summary
• SwitchR: Multi-radio switching architecture– Incrementally deployable
– Energy Savings (72% over WiFi-PSM)
– Can increase battery lifetime substantially
13
Results: VoIP traffic
15
VoIP Benchmark Suite (Using both g711 and g729 codecs)
0
0.2
0.4
0.6
0.8
1
1.2
Ave
rage
Pow
er (N
orm
aliz
ed)
00.2
0.40.6
0.81
1.21.4
1.61.8
2
Ene
rgy-
per-
bit (
Nor
mal
ized
)
Bluetooth-Power WiFi-power Energy-Per-Bit
g711 g729 webg729
Although, bandwidth requirements less than bluetooth channel capacity Web benchmark causes VoIP streams to switch to WiFi
multi-client policy saves upto 65% over cap-dynamic, allows VoIP streams to switch back