![Page 1: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/1.jpg)
MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices
Asaf Cidon*, Tomer M. London*, Sachin Katti, Christos Kozyrakis, Mendel Rosenblum
*Equal contributorsStanford University
![Page 2: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/2.jpg)
New Class of Mobile Applications
October 23, 2011 Slide 2
Augmented Reality
Computer Vision
Motion Sensing
![Page 3: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/3.jpg)
Mobile Client Trends• Mobile CPU performance increasing
– Hitting ‘energy wall’• Can we improve performance and reduce energy
consumption?• Opportunity: network bandwidth increase utilize the cloud
Slide 3October 23, 2011
802.11 Legacy
Mode
802.11b
802.11a
802.11g
802.11n - 40 M
Hz
802.11ac - 80 M
Hz (pro
jection)
1
10
100
1000
Evolution of Wi-Fi Bandwidth
Max
imum
Ban
dwid
th (M
b/s)
![Page 4: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/4.jpg)
Static Client-Server PartitioningDoesn’t Work
• Dynamic resources:– Network bandwidth and latency– Available CPU, memory
• Same code, different platforms:– Smartphones (single-core, multi-core)– Tablets
October 23, 2011 Slide 4
![Page 5: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/5.jpg)
MARS: Adaptive Remote Execution• Opportunistically offload computations to remote
server– Enhance computational capabilities– Decrease energy consumption
• Make dynamic decisions– Adapt to network and CPU variability
October 23, 2011 Slide 5Data CenterMobile Device
![Page 6: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/6.jpg)
Agenda
1. Design of MARS2. Simulator Results and Analysis3. Conclusions
October 23, 2011 Slide 6
![Page 7: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/7.jpg)
Existing Remote Execution Systems
October 23, 2011 Slide 7
The Unit ofRemote Execution
Target of Performance Optimization
RPC
VM
Single-thread application
Multi-threadedapplication
System
CloneCloud [Kirsch et al.,
‘11]
Cloudlets[Satyanarayanan
et al., ‘09]
MAUI [Cuervo et al. ‘10]
Chroma [Balan et al. ‘03]
Odessa [Ra et al. ‘11]
MARS“Cloud-on-
Chip”
![Page 8: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/8.jpg)
Previous Systems:Application Partitioning
October 23, 2011 Slide 8
RPC 1Process 1
RPC 2Process 1
RPC 3Process 1
RPC 4Process 1
RPC 5Process 1
Local Execution Remote Execution
RPC 2Process 3
RPC 1Process 3
RPC 2Process 1
RPC 1Process 2
RPC 1Process 1
RPC Queue
LocalCores
RemoteCores
MARS “Cloud-on-Chip”:System Scheduling
![Page 9: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/9.jpg)
Greedy Algorithm
Slide 9October 23, 2011
Higher POR: better performance gain from offloading
Higher EOR: better energy saving from offloading
PC)NetDelay(Rme(RPC)RemoteExTi
e(RPC)LocalExTimPOR(RPC)
)(RPCrgyNetworkEne
LocalPowere(RPC)LocalExTimEOR(RPC)
EOR ≥ ?
EOR < ?
![Page 10: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/10.jpg)
Remote Server
Local Core
Controller Algorithm
Slide 10October 23, 2011
Priority Queue, sorted by Performance Offload Rank (POR)
Available
Available
EORLocal RemoteBoth
𝟏𝑮
Check EOR Threshold
G (Greediness) trades-off utilization
and energy efficiency
𝑮
RPC 2 (POR 0.4)
RPC 4 (POR 1.3)
RPC 6 (POR 1.8)
RPC 5 (POR 1.9)
RPC 3 (POR 2.5)
RPC 6 (POR 1.8)
![Page 11: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/11.jpg)
Agenda
1. Design of MARS2. Simulator Results and Analysis3. Conclusions
October 23, 2011 Slide 11
![Page 12: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/12.jpg)
Remote Execution Applications
Detection
Recognition
Pic
Barcode
Rendering
Pic
Slide 12
Barcode
Rendering
Pic
Barcode
Rendering
Pic
Detection
Recognition
Pic
Detection
Recognition
Pic
Augmented Reality Face Recognition
![Page 13: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/13.jpg)
Simulator Methodology• Trace-driven simulation• Clients:
– Nokia N900 (single core)– NVIDIA Tegra 250 (multicore)
• Server:– Amazon EC2 Opteron 2007
• Networks:– Outdoors Wi-Fi– Indoors Wi-Fi– 3G
Slide 13June 4, 2011
![Page 14: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/14.jpg)
MARS vs. Static Policies
Slide 14
![Page 15: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/15.jpg)
Nokia N900 Power Consumption
• WiFi: Performance and energy are highly correlated• 3G: trade-off performance and energy
October 23, 2011 Slide 15
Wi-Fi 3GIdle Network Power 1.31 Watts 0.66 Watts
Upload Network Power
1.464 Watts 2.36 Watts
Download Network Power
1.39 Watts 2.26 Watts
Upload Network Power Overhead
10.51% 72.03%
![Page 16: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/16.jpg)
Same Application, Different Networks
Slide 16
![Page 17: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/17.jpg)
Remote Execution with Multicore
Slide 17October 23, 2011
![Page 18: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/18.jpg)
Agenda
1. Design of MARS2. Simulator Results and Analysis3. Conclusions
October 23, 2011 Slide 18
![Page 19: MARS: Adaptive Remote Execution Scheduler for Multithreaded Mobile Devices](https://reader031.vdocument.in/reader031/viewer/2022020417/568143a6550346895db02a7c/html5/thumbnails/19.jpg)
Conclusions
1. Can’t always be greedy– Performance and energy trade-off
2. MARS is optimized for multiple parallel applications and cores
3. MARS “Cloud-on-Chip”: validation of system-level remote execution scheduling– 57% performance increase, 33% energy savings
October 23, 2011 Slide 19