Dynamic Switching of Total-Order Broadcast Protocols

The exchange of total order broadcast protocols for dependable applications has been already studied and solved in several previous papers. To this end, some switching mechanism is needed. All these solutions have been inspired in a two-phase algorithm. The first phase is needed to agree on the need of change, stopping then the old protocol. Whilst the second one is needed to wait for the delivery of all previously broadcast messages, allowing to start later with the new protocol. This introduces a blocking interval that could be quite long in networks with high latency (e.g., WANs) or in systems with slow (i.e., heavily loaded) processes. We propose a new switching mechanism that avoids all such inconveniences, ensuring a fast migration between different broadcast protocols.

[1]  Luís E. T. Rodrigues,et al.  Run-Time Switching Between Total Order Algorithms , 2006, Euro-Par.

[2]  Mark Bickford,et al.  Protocol switching: exploiting meta-properties , 2001, Proceedings 21st International Conference on Distributed Computing Systems Workshops.

[3]  Francesc D. Muñoz-Escoí,et al.  Reducing Transaction Abort Rates with Prioritized Atomic Multicast Protocols , 2008, Euro-Par.

[4]  Dale Skeen,et al.  Nonblocking commit protocols , 1981, SIGMOD '81.

[5]  André Schiper,et al.  Structural and algorithmic issues of dynamic protocol update , 2006, Proceedings 20th IEEE International Parallel & Distributed Processing Symposium.

[6]  Francesc D. Muñoz-Escoí,et al.  On the Cost of Prioritized Atomic Multicast Protocols , 2009, OTM Conferences.

[7]  F. D. Muñoz-Escoí,et al.  Dynamic Total-Order Broadcast Protocol Replacement , 2010 .

[8]  Idit Keidar,et al.  Group communication specifications: a comprehensive study , 2001, CSUR.

[9]  Matti A. Hiltunen,et al.  Constructing adaptive software in distributed systems , 2001, Proceedings 21st International Conference on Distributed Computing Systems.

[10]  Francesc D. Muñoz-Escoí,et al.  Managing Priorities in Atomic Multicast Protocols , 2008, 2008 Third International Conference on Availability, Reliability and Security.

[11]  Fred B. Schneider,et al.  Replication management using the state-machine approach , 1993 .

[12]  André Schiper,et al.  Comparative Performance Analysis of Ordering Strategies in Atomic Broadcast Algorithms , 2003 .

[13]  Pablo Galdámez,et al.  Group communication potocol replacement for high availability and adaptiveness , 2005 .

[14]  Fred B. Schneider,et al.  The primary-backup approach , 1993 .

[15]  Leslie Lamport,et al.  How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs , 2016, IEEE Transactions on Computers.