REVENTS: Facilitating Event-Driven Distributed HPC Applications

Modern scientific applications that need to share geographically scattered resources and dynamically adapt to changes in the environment pose challenges to traditional parallel and distributed programming paradigms. Distributed component frameworks attempt to address the demands of contemporary HPC applications by enabling coarse-grained decomposition and loose coupling. Nonetheless, components usually communicate via synchronous RPC, which is not suitable for interactive applications. This paper introduces a novel distributed event notification system, called REVENTS,which enables both synchronous and decentralized asynchronous component interactions. The REVENTS system is based on a topic-list publisher-subscriber model. It integrates and enhances common technologies for messaging, events, and group communication. The article introduces the REVENTS API, its reference implementation, and its application in the H2O metacomputing framework. Presented experimental results confirm REVENTS' usability in distributed HPC scenarios.

[1]  Karsten Schwan,et al.  Event Services in High Performance Systems , 2001, Cluster Computing.

[2]  Vaidy S. Sunderam,et al.  RMIX: a multiprotocol RMI framework for Java , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[3]  Yi Huang,et al.  A comparative study of Web services-based event notification specifications , 2006, 2006 International Conference on Parallel Processing Workshops (ICPPW'06).

[4]  Ralph Johnson,et al.  design patterns elements of reusable object oriented software , 2019 .

[5]  Rahul Sharma,et al.  Java¿ Message Service API Tutorial and Reference: Messaging for the J2EE¿ Platform , 2002 .

[6]  Vaidy S. Sunderam,et al.  Performance and client heterogeneity in service-based metacomputing , 2004, 18th International Parallel and Distributed Processing Symposium, 2004. Proceedings..

[7]  Vaidy S. Sunderam,et al.  Towards Self-Organizing Distributed Computing Frameworks: The H2O Approach , 2003, Parallel Process. Lett..

[8]  Yi Huang,et al.  WS-Messenger: a Web services-based messaging system for service-oriented grid computing , 2006, Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06).