Large scale dynamic distributed systems

Working with Nancy Lynch and Idit Keidar we have studied large scale dynamic distributed systems in which processes can join, leave and fail. We have developed a number of algorithms to solve Atomic Broadcast in such systems, under different network assumptions. In particular, we have developed an algorithm for early-delivery atomic broadcast in such network. I have also worked with Prof. Michael Ben-Or on lower bounds for synchronous randomized consensus. Check the papers below for more details.

Related publications

  • Z. Bar-Joseph, I. Keidar, and N. Lynch
    Early-Delivery Dynamic Atomic Broadcast.
    16th International Symposium on DIStributed Computing (DISC), pp 1-16, 2002.

  • Z. Bar-Joseph, I. Keidar, T. Anker, and N. Lynch:
    QoS Preserving Totally Ordered Multicast.
    In the 5th International Conference On Principles Of DIstributed Systems (OPODIS), pages 143-162, Paris, France, December, 2000.
    Previous version: MIT Technical Report MIT-LCS-TR-796, January 2000.
  • Z. Bar-Joseph and M. Ben-Or
    A Tight Lower Bound for Randomized Synchronous Consensus
    in proceedings of the Seventeenth Annual ACM Symposium on Principles of Distributed Computing (PODC), June 28 - July 2 1998, pp. 193-199, Puerto Vallarta, Mexico.
    Best student paper award