The computational power of population protocols

We consider the model of population protocols introduced by Angluin et al. (Computation in networks of passively mobile finite-state sensors, pp. 290–299. ACM, New York, 2004), in which anonymous finite-state agents stably compute a predicate of the multiset of their inputs via two-way interactions in the family of all-pairs communication networks. We prove that all predicates stably computable in this model (and certain generalizations of it) are semilinear, answering a central open question about the power of the model. Removing the assumption of two-way interaction, we also consider several variants of the model in which agents communicate by anonymous message-passing where the recipient of each message is chosen by an adversary and the sender is not identified to the recipient. These one-way models are distinguished by whether messages are delivered immediately or after a delay, whether a sender can record that it has sent a message, and whether a recipient can queue incoming messages, refusing to accept new messages until it has had a chance to send out messages of its own. We characterize the classes of predicates stably computable in each of these one-way models using natural subclasses of the semilinear predicates.

[1]  Arvin Park,et al.  The Processor Identity Problem , 1990, Inf. Process. Lett..

[2]  Graham Higman,et al.  Ordering by Divisibility in Abstract Algebras , 1952 .

[3]  Michael J. Fischer,et al.  A simple game for the study of trust in distributed systems , 2009, Wuhan University Journal of Natural Sciences.

[4]  Sebastiano Vigna,et al.  An Effective Characterization of Computability in Anonymous Networks , 2001, DISC.

[5]  Rachid Guerraoui,et al.  Anonymous and fault-tolerant shared-memory computing , 2007, Distributed Computing.

[6]  James Aspnes,et al.  Wait-free consensus with infinite arrivals , 2002, STOC '02.

[7]  John E. Hopcroft,et al.  On the Reachability Problem for 5-Dimensional Vector Addition Systems , 1976, Theor. Comput. Sci..

[8]  Marina Papatriantafilou,et al.  Randomized naming using wait-free shared variables , 1998, Distributed Computing.

[9]  Ambuj K. Singh,et al.  Naming symmetric processes using shared variables , 2005, Distributed Computing.

[10]  Sebastiano Vigna,et al.  Computing anonymously with arbitrary knowledge , 1999, PODC '99.

[11]  Michael J. Fischer,et al.  Computation in networks of passively mobile finite-state sensors , 2004, PODC '04.

[12]  Hong Jiang,et al.  Distributed systems of simple interacting agents , 2007 .

[13]  L. Dickson Finiteness of the Odd Perfect and Primitive Abundant Numbers with n Distinct Prime Factors , 1913 .

[14]  Dan ie l T. Gil lespie A rigorous derivation of the chemical master equation , 1992 .

[15]  Boaz Patt-Shamir,et al.  The Las-Vegas Processor Identity Problem (How and When to Be Unique) , 2000, J. Algorithms.

[16]  Michael J. Fischer,et al.  Stabilizing Consensus in Mobile Networks , 2006, DCOSS.

[17]  Shang-Hua Teng Space Efficient Processor Identity Protocol , 1990, Inf. Process. Lett..

[18]  Michael A. Gibson,et al.  Efficient Exact Stochastic Simulation of Chemical Systems with Many Species and Many Channels , 2000 .

[19]  David Eisenstat,et al.  Stably computable predicates are semilinear , 2006, PODC '06.

[20]  R. Kannan,et al.  Convex Sets and their Applications , 2006 .

[21]  Michael J. Fischer,et al.  Self-stabilizing population protocols , 2005, TAAS.

[22]  J. Berstel,et al.  Context-free languages , 1993, SIGA.

[23]  David Eisenstat,et al.  Fast computation by population protocols with a leader , 2006, Distributed Computing.

[24]  Sam Toueg,et al.  Wakeup under Read/Write Atomicity , 1990, WDAG.

[25]  P. McMullen Convex Sets and Their Applications , 1982 .

[26]  David Eisenstat,et al.  On the Power of Anonymous One-Way Communication , 2005, OPODIS.

[27]  Michael J. Fischer,et al.  Self-stabilizing Leader Election in Networks of Finite-State Anonymous Agents , 2006, OPODIS.

[28]  Oscar H. Ibarra,et al.  Catalytic P systems, semilinear sets, and vector addition systems , 2004, Theor. Comput. Sci..

[29]  Faith Ellen,et al.  Hundreds of impossibility results for distributed computing , 2003, Distributed Computing.

[30]  Hagit Attiya,et al.  Computing in Totally Anonymous Asynchronous Shared Memory Systems , 1998, DISC.

[31]  H IbarraOscar,et al.  Catalytic P systems, semilinear sets, and vector addition systems , 2004 .

[32]  Dana Angluin,et al.  Local and global properties in networks of processors (Extended Abstract) , 1980, STOC '80.

[33]  Naoshi Sakamoto,et al.  Comparison of initial conditions for distributed algorithms on anonymous networks , 1999, PODC '99.

[34]  Alessandro Panconesi,et al.  On the importance of having an identity or, is consensus really universal? , 2005, Distributed Computing.

[35]  Dana Angluin,et al.  Urn Automata , 2003 .

[36]  Norman T. J. Bailey,et al.  The Mathematical Theory of Infectious Diseases , 1975 .

[37]  Rachid Guerraoui,et al.  When Birds Die: Making Population Protocols Fault-Tolerant , 2006, DCOSS.

[38]  Michael J. Fischer,et al.  Stably Computable Properties of Network Graphs , 2005, DCOSS.

[39]  Rachid Guerraoui,et al.  What Can Be Implemented Anonymously? , 2005, DISC.