Now showing items 2-12 of 12

    • Failure Detectors in Omission Failure Environments 

      Dolev, Danny; Friedman, Roy; Keidar, Idit; Malkhi, Dahlia (Cornell University, 1996-09)
      We study failure detectors in an asynchronous environment that admits message omission failures. In such environments, processes may fail by crashing, but may also {\em disconnect\/} from each other. We adapt Chandra and ...
    • The Hierarchical Daisy Architecture for Causal Delivery 

      Baldoni, Roberto; Friedman, Roy; van Renesse, Robbert (Cornell University, 1996-09)
      In this paper, we propose the {\em hierarchical daisy architecture}\/, which provides causal delivery of messages sent to any subset of processes. The architecture provides fault tolerance and maintains the amount of ...
    • High-Performance Replicated Distributed Objects in PartitionableEnvironments 

      Friedman, Roy; Vaysburd, Alexey (Cornell University, 1997-07)
      This paper presents an implementation of replicated distributed objects in asynchronous environments prone to node failures and network partitions. This implementation has several appealing properties: It guarantees that ...
    • Implementing Replicated State Machines Over Partitionable Networks 

      Friedman, Roy; Vaysburd, Alexey (Cornell University, 1996-04)
      This paper presents an implementation of a replicated state machine in asynchronous distributed environments prone to node failures and network partitions. This implementation has two appealing properties: It allows minority ...
    • Load Balancing Schemes for High-Throughput Distributed Fault-TolerantServers 

      Friedman, Roy; Mosse, Daniel (Cornell University, 1996-12)
      Clusters of workstations, connected by a fast network, are emerging as a viable architecture for building high-throughput fault-tolerant servers. This type of architecture is more scaleable and more cost-effective than a ...
    • The Maestro Group Manager: A Structuring Tool For Applications WithMultiple Quality of Service Requirements 

      Birman, Ken; Friedman, Roy; Hayden, Mark (Cornell University, 1997-02)
      {\em Maestro} is a tool for managing sets of protocol stacks that satisfy varied quality of service or security requirements. Intended primarily for multimedia groupware settings, it permits a single application to efficiently ...
    • Packing Messages as a Tool for Boosting the Performance of TotalOrdering Protocols 

      Friedman, Roy; van Renesse, Robbert (Cornell University, 1995-07)
      This paper compares the throughput and latency of four protocols that provide total ordering. Two of these protocols are measured with and without message packing. We used a technique that buffers application messages for ...
    • Strong and Weak Virtual Synchrony in Horus 

      Friedman, Roy; van Renesse, Robbert (Cornell University, 1995-08)
      A formal definition of {\em strong virtual synchrony}, capturing the semantics of virtual synchrony as implemented in Horus, is presented. This definition has the nice property that every message is delivered within the ...
    • Trading Consistency for Availability in Distributed Systems 

      Friedman, Roy; Birman, Ken (Cornell University, 1996-04)
      This paper shows that two important classes of actions, {\em non left commuting}\/ and {\em strongly non commuting}, cannot be executed by concurrent partitions in a system that provides serializable services. This result ...
    • Using Group Communication Technology to Implement a Reliable andScalable Distributed IN Coprocessor 

      Friedman, Roy; Birman, Ken (Cornell University, 1996-08)
      In this paper we explore the use of group communication technology, developed in the Horus project to implement a reliable and scalable distributed IN coprocessor. The proposed implementation can handle up to 20,000 calls ...
    • Using Virtual Synchrony to Develop Efficient Fault TolerantDistributed Shared Memories 

      Friedman, Roy (Cornell University, 1995-03)
      This paper shows how to define consistency conditions for distributed shared memories in virtually synchronous environments. Such definitions allow to develop fault tolerant implementations of distributed shared memories, ...