MDC-Cast: A Total-Order Broadcast Protocol for Multi-Datacenter Environments

The recent Total-Order Broadcast protocols that have been designed to sustain high throughput and low latency target fully switched environments, such as small datacenters and clusters. These protocols fail to achieve good performance in multi-datacenter environments, that are characterized by non-uniform network connectivity among a set of remote datacenters. More precisely, machines within a datacenter are connected using a fully switched network, whereas machines across datacenters use shared inter-datacenter network cables. This paper presents a novel Total-Order Broadcast protocol, called MDC-cast that specifically targets multi-datacenter environments.

[1]  Xiaohua Jia.  A Total Ordering Multicast Protocol Using Propagation Trees , 1995, IEEE Trans. Parallel Distributed Syst..

[2]  Jean-Yves Le Boudec,et al.  Rate adaptation, Congestion Control and Fairness: A Tutorial , 2000 .

[3]  Fernando Pedone,et al.  Rethinking State-Machine Replication for Parallelism , 2013, 2014 IEEE 34th International Conference on Distributed Computing Systems.

[4]  Jo-Mei Chang,et al.  Reliable broadcast protocols , 1984, TOCS.

[5]  Hein Meling,et al.  Jgroup/ARM: a distributed object group platform with autonomous replication management , 2008, Softw. Pract. Exp..

[6]  Katherine Guo,et al.  Moving the ensemble communication system to NT and Wolfpack , 1997 .

[7]  Kenneth P. Birman,et al.  Performance of the ISIS Distributed Computing Toolkit , 1994 .

[8]  Robbert van Renesse,et al.  Horus: a flexible group communication system , 1996, CACM.

[9]  Vivien Quéma,et al.  FastCast: A Throughput- and Latency-Efficient Total Order Broadcast Protocol , 2013, Middleware.

[10]  Francisco Moura,et al.  Optimistic total order in wide area networks , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..

[11]  Robbert van Renesse,et al.  Reliable Distributed Computing with the Isis Toolkit , 1994 .

[12]  Danny Dolev,et al.  On the minimal synchronism needed for distributed consensus , 1983, 24th Annual Symposium on Foundations of Computer Science (sfcs 1983).

[13]  Keith Marzullo,et al.  Mencius: Building Efficient Replicated State Machine for WANs , 2008, OSDI.

[14]  David G. Andersen,et al.  There is more consensus in Egalitarian parliaments , 2013, SOSP.

[15]  Fernando Pedone,et al.  Ridge: High-Throughput, Low-Latency Atomic Multicast , 2015, 2015 IEEE 34th Symposium on Reliable Distributed Systems (SRDS).

[16]  Fernando Pedone,et al.  Multi-Ring Paxos , 2012, IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2012).

[17]  Lau Cheuk Lung,et al.  BAMcast - byzantine fault-tolerant consensus service for Atomic Multicast in large-scale networks , 2013, 2013 IEEE Symposium on Computers and Communications (ISCC).

[18]  Amar Phanishayee,et al.  PLATO: Predictive Latency-Aware Total Ordering , 2006, 2006 25th IEEE Symposium on Reliable Distributed Systems (SRDS'06).

[19]  Leslie Lamport,et al.  The part-time parliament , 1998, TOCS.

[20]  Jongsung Kim,et al.  A total ordering protocol using a dynamic token-passing scheme , 1997, Distributed Syst. Eng..

[21]  Kenneth P. Birman,et al.  Reliable communication in the presence of failures , 1987, TOCS.

[22]  Nancy A. Lynch,et al.  Impossibility of distributed consensus with one faulty process , 1985, JACM.

[23]  Sam Toueg,et al.  Unreliable failure detectors for reliable distributed systems , 1996, JACM.

[24]  Claudiu Danilov,et al.  The Spread Toolkit: Architecture and Performance , 2004 .

[25]  Kenneth P. Birman,et al.  Exploiting virtual synchrony in distributed systems , 1987, SOSP '87.

[26]  Rachid Guerraoui,et al.  Throughput optimal total order broadcast for cluster environments , 2010, TOCS.

[27]  Flaviu Cristian,et al.  High-performance asynchronous atomic broadcast , 1997, Distributed Syst. Eng..

[28]  Fernando Pedone,et al.  Ring Paxos: A high-throughput atomic broadcast protocol , 2010, 2010 IEEE/IFIP International Conference on Dependable Systems & Networks (DSN).