ORGs for Scalable, Robust, Privacy-Friendly Client Cloud Computing

Many-core Client Cloud Computing

Many-core is a computer architecture in which many processors (called “cores”) are printed on a chip and the chips in turn can be physically connected using a multichip package. The cores have extremely high-bandwidth low-latency communications often using shared coherent cache memory. Chips from Intel and AMD currently have up to 4 cores on a chip. Industry projections are that follow-on chips will soon have dozens of cores.

Cloud Computing is a paradigm in which information is permanently stored in servers on the Internet and cached temporarily on clients that include desktops, entertainment centers, table computers, notebooks, wall computers, handhelds, etc.

Many-core stands to transform Cloud Computing on the client in the following ways:

1. Extending the Cloud into the client. Silverlight^[2] (Microsoft), Android^[1] (Google), AIR (Adobe), and Java (Sun) will extend Cloud Computing into the client. Thus many of the issues that must currently be addressed on the Internet will now have to be addressed on each client!
2. Guaranteeing the privacy of client information in Cloud Computing. Many-core will make it possible to maintain client information encrypted on servers so that it can be decrypted only by using the client’s private key. (The information is unencrypted only when cached on a client.)
3. Scalable Semantic Integration. Client cloud computing will provide greater integration of user information by obtaining it from the servers of competing vendors without requiring them to interact with each other. Also semantic integration will support better advertising relevance and targeting without exposing client privacy.

Without many-core, the above transformation would be impossible because there would be insufficient processing power on the client. Industry projections are that cycle frequencies on chips will increase only very gradually over the next decade. Consequently, many-core is the only way to obtain the requisite processing power on the client for the user interfaces and applications required to extend Cloud Computing into the client.

Of course, the above developments will require rather sophisticated software systems. By what principles and methods will these systems be developed?

For the continuation of the published article, please see the following:

Summary

Acknowledgments

At the 2006 American Association of Artificial Intelligence (AAAI) Spring Symposium, Ed Feigenbaum, Mehmet Göker, Doug Lenat, Dan Shapiro, and others provided valuable feedback. Fanya S. Montalvo made extensive detailed comments and suggestions for improvement. Dan Shapiro and Mehmet Göker suggested revising an earlier version of this paper and subsequently Dan provided extensive help in getting started and editing. Pablo Noriega, Munindar Singh, Liz Sonenberg, and Hiroki Suguri participated in a panel chaired by the author at the 2007 Autonomous Agents and Multi-Agent Systems (AAMAS) that discussed some of the challenges presented in this paper. At AAAI’08, conversations with Bruce Buchanan, Mehmet Göker, Ben Kuipers, Erik Sandewall, Dan Shapiro, Reid Smith, and others were very helpful. The Stanford Logic group led by Michael Genesereth has provided a supportive environment for the further development of some of the ideas. Recently, Fanya S. Montalvo, Charles Petrie, and Steve Woods made valuable suggestions for improvement.

References

    1.  C. Hewitt and J. Inman, “DAI Betwixt and Between:  From ‘Intelligent Agents’ to Open Systems Science,” IEEE Transactions on Systems, Man, and Cybernetics, Nov. /Dec. 1991, pp. 1409-1419.

    2.  C. Hewitt, “What is Commitment? Physical, Organizational, and Social,” Coordination, Organizations, Institutions and Norms in Multi-Agent Systems II, P. Noriega et al., eds., Springer Verlag. 2007, pp. 293-307.

    3.  W. Kornfeld and C. Hewitt, “The Scientific Community Metaphor,” IEEE Trans. Systems, Man, and Cybernetics, Jan. 1981, pp 24-33.

    4.  Social Science Research, Technical Systems and Cooperative Work, G. Bowker, S. L. Star, W. Turner, and L. Gasser, eds., Lawrence Earlbaum, 1997.

    5.  P. Noriega, Agent Mediated Auctions: The Fishmarket Metaphor, doctoral dissertation, Univ. of Barcelona, 1997.

    6.  M. Singh and M. Huhns, Service-Oriented Computing: Semantics, Processes, Agents, John Wiley & Sons, 2005.

    7.  B. Horling and V. Lesser, “Using ODML to Model Multi-Agent Organizations,” Proc. International Conference on Intelligent Agent Technology (IAT’05) IEEE Computer Society. 2005, pp. 72-80.

    8.  P. Neumann, “Holistic Approaches to Trustworthiness, Security, & Privacy,” Cybersecurity Summit 2008 for NSF Large Research Facilities 7 May 2008;

    9.  C. Hewitt (2008a), Norms and Commitment for ORGs (Organizations of Restricted GeneralityTM): Strong Paraconsistency and Participatory Behavioral Model Checking

  10.  R. Kowalski and M. Sergot, “A Logic-based Calculus of Events,” New Generation Computing, February 1986. Ohmsha, Ltd., pp. 67-94.

  11.  D. Bianculli, A. Morzenti, M. Pradella, P. San Pietro, and P. Spoletini, “Trio2Promela: A Model Checker for Temporal Metric Specifications,” IEEE International Conference on Software Engineering (ICSE’07), 2007, pp 61-62.

  12.  N. Desai, Z. Cheng, A. Chopra, and M. Singh, “Toward Verification of Commitment Protocols and their Compositions,” Autonomous Agents and Multi-Agent Systems (AAMAS’07). ACM. 2007. pp 144-146.

  13.  M. Kassoff, LM Zen, A. Garg, and M. Genesereth, “PrediCalc: A Logical Spreadsheet Management System,” Proc. 31st Int’l Conf. Very Large Databases (VLDB), VLDB Endowment. 2005, pp. 1247-1250.

  14.  C. Hewitt (2008b), “Large-scale Organizational Computing requires Unstratified Reflection and Strong Paraconsistency,” Coordination, Organizations, Institutions, and Norms in Agent Systems III, J. Sichman et al., eds., Springer-Verlag, 2008, pp 110-124.

  15.  D. Fensel and F. van Harmelen, “Unifying Reasoning and Search to Web Scale,” IEEE Internet Computing, Mar./Apr. 2007, pp. 93-95.

  16.  C. Hewitt (2008c), “Common sense for concurrency and strong paraconsistency using unstratified inference and reflection” ArXiv. December 30, 2008.

  17.  M. Schwartz, “The Rise of Malwebolence,” International Herald Tribune, August 1, 2008.

  18.  M. Callon, “Some elements of a sociology of translation: domestication of the scallops and fishermen of St. Brieuc Bay,” in Power, Action and Belief edited by John Law, Routledge & Kegan Paul, 1986.  pp. 196-233.

  19.  P. Correa, A. Correa and M. Askanas, Special Report: Internet encyclopaedias go head to head December 15, 2005.

  20.  C. Dupont and M. McNally, “Knol is open to everyone,” Official Google Blog.