Optimizing the Energy Efficiency of Message Exchanging for Service Distribution in Interoperable Infrastructures
Authors:
- Nik BessisNik Bessis
- Stelios SotiriadisSSStelios Sotiriadis
- Florin PopFPFlorin Pop
- Valentin CristeaVCValentin Cristea
- 1 September 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 105-112
Abstract
Optimizing the competence of message exchanging algorithm in large-scale grids and inter-clouds could be proven to be a crucial factor in achieving an efficient resource management. Traditional solutions usually incorporate either probabilistic or flooding message exchanging techniques mainly due to the dynamic and unpredictable formation of the infrastructure. Our vision encompasses a total decentralized nodes topology in which message exchanging algorithm allows dissemination of communication messages within a decoupled node formation setting. We consider various e-infrastructure nodes that exchange simple messages with linking nodes regarding resource competence for executing certain service(s) requirements. The optimization criterion is to improve the energy efficiency of the network performance for message exchanging in two cases as follows. Firstly, a requester sends messages to all interconnected nodes and gets messages only from resources available to execute it. Secondly, the requester sends one message for all of the jobs of its local pool and gets a respond from available nodes, and then obtainable resources are ranked and hierarchically categorized based on the performance criterion e.g. latency competency. The algorithm is simulated and results obtained are very supportive.Keywords
This publication has 12 references indexed in Scilit:
- Advancing Inter-cloud Resource Discovery Based on Past Service Experiences of Transient Resource Clustering2012
- Decentralized meta-brokers for inter-cloud: Modeling brokering coordinators for interoperable resource management2012
- Meta-scheduling issues in interoperable HPCs, grids and cloudsInternational Journal of Web and Grid Services, 2012
- Modelling Requirements for Enabling Meta-scheduling in Inter-Clouds and Inter-Enterprises2011
- Graph Theory2008
- An efficient MPI_allgather for grids2007
- Total Exchange Performance Prediction on Grid Environments2007
- Collective communication on architectures that support simultaneous communication over multiple links2006
- DEMEM: Distributed Evidence-Driven Message Exchange Intrusion Detection Model for MANET2006
- Network performance-aware collective communication for clustered wide-area systemsParallel Computing, 2001