SwitchR: Lessening Framework Power Utilization in a Multi-Customer Multi-Radio Environment.

Uploaded on:
Category: Fashion / Beauty
Wearable/Mobile Computers Power Consumption is essential! ... Considers movement forced by different gadgets in a multi-customer situation. Exchanging choice ...
Slide 1

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)

Slide 2

Wearable and Mobile Devices: Increasing Functionality Faster processors, more memory Applications are progressively correspondence concentrated Streaming video, VoIP, Downloading records Multiple remote radios regularly incorporated on single gadget (Bluetooth for PANs, WiFi for high-transfer speed information access) Wearable/Mobile Computers  Power Consumption is vital! Restricted by battery lifetime Communication over WiFi diminishes battery lifetime much further… . Sometimes up to half of aggregate vitality channel!

Slide 3

Reducing the vitality for correspondence Opportunity: Availability of numerous radio interfaces … Can all be utilized for information exchange Different attributes : transmission capacity, range, power utilization Typically work as confined frameworks, Can we facilitate use to give a bound together system association ? Consistently switch between radios Primary Goal: Save vitality X +

Slide 4

Radio Characteristics Higher throughput radios have a lower vitality/bit esteem … have a higher unmoving force utilization … and they have diverse reach qualities

Slide 5

Multi-Radio Switching CoolSpots [Mobisys \'06] : Multi-Radio exchanging for a solitary customer situation Specialized access point (Bluetooth + WiFi) Switching choices – Local to customer SwitchR : Leverage existing WiFi APs : Incrementally deployable Considers activity forced by different gadgets in a multi-customer situation Switching choice – worldwide since it influence different customers Evaluate vitality investment funds on a circulated testbed Problem Statement: Reduce vitality utilization by picking suitable radio interface, while mulling over different customers.

Slide 6

SwitchR Architecture Infrastructure Network BTG (Bluetooth Gateway) Bluetooth Link MD1 WiFi Link Ethernet Link Wi-Fi Zone MD2 Wi-Fi AP (WFAP) MD3 MD = Mobile Devices MD4 Switching Policy: Hybrid Approach Application prerequisites at hubs (neighborhood) Channel quality and transmission capacity (worldwide) Switching Mechanism: Network Level Reconfigurations ARPs and Routing upgrades

Slide 7

Multi-Client Switching Policy Hybrid way to deal with settle on exchanging choices Local learning (hub level) Global (channel use by different hubs) Switching up (Bluetooth  WiFi) ICMP reaction time and radio RSSI values Capture application needs and channel attributes Switching-down (WiFi  Bluetooth) Measure application data transmission necessities Periodically question BTG for remaining limit Measure channel/join quality (nearby)

Slide 8

Evaluation: Testbed BTG (Bluetooth Gateway) Infrastructure Network Bluetooth (Always Connected) MD1 WiFi (Dynamically Switched) Static Wired Connection Wi-Fi Zone MD2 Wi-Fi AP MD3 Mobile Device (MD) MD4 Stargate2 hub Stargate2 research stage WiFi + Bluetooth + Integrating force and information observing Benchmark applications are striped crosswise over gadgets

Slide 9

Evaluation: Benchmarks Baselines: Idle: associated, yet no information exchange Transfer: mass TCP information exchange Streaming: Media: 128k, 156k and g711 VoIP codec Various QoS necessities Web: Combination of unmoving and information exchange Idle: "think time" Small exchange: essential website pages Bulk exchange: reports or media

Slide 10

Evaluation: Switching Policies Baselines arrangements "Wifi-CAM" (Awake Mode) "Wifi-PSM" (Power Save Mode) S ingle-Client based " top element" exchanging strategy SwitchR: "multi-customer" exchanging approach Combines both nearby (per customer) and worldwide learning

Slide 11

Results: Baselines Switching strategies perform better that WiFi strategies for "unmoving" benchmark, comparable for "exchange"

Slide 12

Results: multi-customer strategy sets aside to 62% over single-customer top element arrangement VoIP and spilling benchmarks advantage most since streams can utilize BT channel

Slide 13

Summary SwitchR: Multi-radio exchanging engineering Incrementally deployable Energy Savings (72% over WiFi-PSM) Can build battery lifetime considerably

Slide 14

Thank You! Site : http://mesl.ucsd.edu/yuvraj Email : yuvraj@cs.ucsd.edu

Slide 15

Results: VoIP activity Although, data transmission prerequisites not exactly bluetooth channel limit Web benchmark causes VoIP streams to change to WiFi multi-customer approach spares upto 65% over top element, permits VoIP streams to switch back

View more...