D1HT: a distributed one hop hash table
IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
Integrating local job scheduler – LSFTM with GfarmTM
ISPA'05 Proceedings of the Third international conference on Parallel and Distributed Processing and Applications
Implementation of cooperative cognition under web environment
ISPA'05 Proceedings of the 2005 international conference on Parallel and Distributed Processing and Applications
Self-star Properties in Complex Information Systems
| Hi-index | 0.00 |
We present the framework of a new grid architecturebased on the peer-to-peer and the component paradigms.In our architecture, several peer-to-peer components areloosely coupled or even independent from each other. Componentsaffect others indirectly by monitoring the system,recognizing changes caused by other components' actions,and by issuing own actions. This approach is generic andopen for future extensions. It allows to design scalable, self-optimizing,and resilient grid systems that have no singlepoint of failure.A component is a distributed software system that usesa peer-to-peer algorithm to achieve its goal. The peer-to-peerparadigm was invented for large, voluntary PC networks.We apply this concept to the field of self-optimizinggrid systems and show how this new architecture can bedeployed in the domain of data management. Here, variousmanagement aspects like replica creation, placement,access optimization, synchronization, etc. are cooperativelysolved in separate peer-to-peer component layers. All componentsrun concurrently. Each of them optimizes the systemwith respect to its goals. This separate optimizationprocess performed by the interaction of small, manageablecomponents yields similar or even better results, while beingless complex than the holistic approach that tries to optimizethe system in one step.