1 / 30

Computing for the Future of the Planet (at Scale) Andy Hopper

Computing for the Future of the Planet (at Scale) Andy Hopper. A. Hopper and A. Rice, “Computing for the Future of the Planet”, Phil. Trans. R. Soc. A , Oct 2008. Google Tech Talk Video, 14 May 2008 Other papers at: www.cl.cam.ac.uk/research/dtg/research/wiki/CFTFP.

azalia-kemp
Download Presentation

Computing for the Future of the Planet (at Scale) Andy Hopper

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Computing for theFuture of the Planet(at Scale)Andy Hopper A. Hopper and A. Rice, “Computing for the Future of the Planet”, Phil. Trans. R. Soc. A, Oct 2008. Google Tech Talk Video, 14 May 2008 Other papers at: www.cl.cam.ac.uk/research/dtg/research/wiki/CFTFP The Computer Laboratory University of Cambridge

  2. CFP Framework • Optimal Digital Infrastructure • Sense and Optimise • Predict and React • Digital Alternatives to Physical Activities

  3. 1 – Optimal Digital Infrastructure • Energy proportional computing • Virtual machine migration enables energy proportional computing

  4. Use Renewable Energy • Locate data centres directly next to power source • Use network to move jobs to data centre • Maintain service level agreements

  5. Chase Surplus Energy Around the Globe Sun Siemens press picture • Keep moving computing tasks to where energy is available • Use energy that cannot be used for another purpose • At what granularity should jobs be shipped? • Do we ship program, data, or both?

  6. Migration Prediction S. Akoush, R. Sohan, A. Rice • AVG Algorithm: Average of number of memory pages changed per unit time • (MT = Migration Time, DT = Downtime)

  7. The Overall Goal (at Scale) • Optimal Digital Infrastructure • Components switched off if not doing useful work • Energy proportional computing and communications at many levels • Use of energy that is not suitable for other purposes • Components • Servers / Server Farms • Networks • Workstations • Terminals • For the first time over-provisioning may not save the day!

  8. 2 - Sense and Optimise • A sensor-based digital model of the planet • “Googling” Earth! • “Googling” Space-Time! • “The Google of Things” • How do we do it? • coverage • fidelity • scalability • performance • usefulness

  9. Future Street View – Heat Sensing?

  10. A Better Shoe R. Harle + = Insole Augmented with Force Sensitive Resistors Custom wireless sensor node Insole Augmented with Force Sensitive Resistors

  11. 60 Meter Sprint Acceleration Deceleration Speed Maintenance

  12. Speed of Runner Each colour represents data from a different person Speed (m/s) Contact Time (s)

  13. Anomalies

  14. Sensor Fusion and Fast Data Analysis Synchronised foot pressure data 205fps video

  15. Location in Industry - BMW Plant Installation facts: 1.7 km line, Sensors: 350, Tags: 1,000, Accuracy: 30 cm in 3D, Events/day: 150,000, Reliability: 99.99%

  16. VBL Bus Depot Luzern Installation facts: Area: 15,000 m² in 1 depot | Sensors: 48 | Accuracy: 100 cm in 2D | Reliability: 99.9%

  17. Video

  18. Precise Location Sensing at Scale • Diffusion of hardware into conventional machines and devices • System is cellular and scaleable at all levels

  19. Global Personal Energy Meter - PEM S. Hay, A. Rice • Complete • all energy accounted for: sensed, embedded, shared, hypothecated • Accurate / Bounded / Personalised • my actions relate to me only • Sensible • incentives work correctly • Trustworthy • rules are understood: reciprocity, availability • fidelity / error bounds • security / privacy: “bad” things cannot happen 21

  20. PEM Implementation S. Hay, A. Rice • Information about an individuals energy consumption • measure, interpret, postulate, allocate • Use World Model • crowdsource data • upload own energy use to help global optimisation • download energy profile of devices, goods, physical places • Apportion energy • to individual, group, thing, place • Lots of lovely computing problems! • measurement, indexing, caching, event-delivery, prediction, use of social networking, security, privacy, correctness, …

  21. What are the rules for apportionment? Allocate equal share of total load independent of use Allocate unequal share of total load, eg to current occupants only Allocate equal share of base load, but my incremental load Power consumption and occupancy

  22. Apportionment for transportation systems? • Walking • my food intake? • Car • equivalent to office? • Bus • equivalent to building? • apportion the cost of a bus service over all the passengers each day? • All public transport ? • What methods / policies / principles will be acceptable? 24

  23. OpenRoomMap: crowd-sourced maps of buildings A. Rice, O. Woodman • Association of objects to places (and individuals) • Useful and rich dataset

  24. Personal Energy Meter: Android App D. Piggot, A. Beresford, A. Rice

  25. 3 - Predict and React • More issues to come!

  26. 4 - Developing World: Digital Platform Capacity Fibre links to the world Wireless networks 3G/EDGE WiMax Femtocells Cheap, commodity hardware: (Smart)phones ($5) Netbooks ($70) Low-cost applications

  27. How can-might-should-will this evolve?

  28. What has been achieved since 1990?

More Related