The case for crowd computing

We introduce and motivate "crowd computing", which combines mobile devices and social interactions to achieve large-scale distributed computation. An opportunistic network of mobile devices offers substantial aggregate bandwidth and processing power. In this paper, we analyse encounter traces to place an upper bound on the amount of computation that is possible in such networks. We also investigate a practical task-farming algorithm that approaches this upper bound, and show that exploiting social structure can dramatically increase its performance.

[1]  Rajkumar Buyya,et al.  Parallel Programming Models and Paradigms , 1998 .

[2]  David Tse,et al.  Mobility increases the capacity of ad-hoc wireless networks , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[3]  Anders Lindgren,et al.  Probabilistic routing in intermittently connected networks , 2003, MOCO.

[4]  Deborah Estrin,et al.  Directed diffusion for wireless sensor networking , 2003, TNET.

[5]  Matt Welsh,et al.  Programming Sensor Networks Using Abstract Regions , 2004, NSDI.

[6]  Anders Lindgren,et al.  Probabilistic Routing in Intermittently Connected Networks , 2004, SAPIR.

[7]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[8]  David P. Anderson,et al.  BOINC: a system for public-resource computing and storage , 2004, Fifth IEEE/ACM International Workshop on Grid Computing.

[9]  Douglas Thain,et al.  Distributed computing in practice: the Condor experience , 2005, Concurr. Pract. Exp..

[10]  Alex Pentland,et al.  Reality mining: sensing complex social systems , 2006, Personal and Ubiquitous Computing.

[11]  Marco Conti,et al.  Opportunistic networking: data forwarding in disconnected mobile ad hoc networks , 2006, IEEE Communications Magazine.

[12]  Yuan Yu,et al.  Dryad: distributed data-parallel programs from sequential building blocks , 2007, EuroSys '07.

[13]  Mads Haahr,et al.  Social network analysis for routing in disconnected delay-tolerant MANETs , 2007, MobiHoc '07.

[14]  Eiko Yoneki,et al.  Visualizing communities and centralities from encounter traces , 2008, CHANTS '08.

[15]  Wen-mei W. Hwu,et al.  Optimization principles and application performance evaluation of a multithreaded GPU using CUDA , 2008, PPoPP.

[16]  Steven Hand,et al.  Spread-Spectrum Computation , 2008, HotDep.

[17]  Jon Crowcroft,et al.  D3N: programming distributed computationin pocket switched networks , 2009, MobiHeld '09.

[18]  Eugene Marinelli,et al.  Hyrax: Cloud Computing on Mobile Devices using MapReduce , 2009 .

[19]  Steven Hand,et al.  Scripting the Cloud with Skywriting , 2010, HotCloud.

[20]  Pan Hui,et al.  BUBBLE Rap: Social-Based Forwarding in Delay-Tolerant Networks , 2008, IEEE Transactions on Mobile Computing.