Adaptive load sharing in homogeneous distributed systems
IEEE Transactions on Software Engineering
Deciding which queue to join: Some counterexamples
Operations Research
A comparison of receiver-initiated and sender-initiated adaptive load sharing
Performance Evaluation
Analysis of the Effects of Delays on Load Sharing
IEEE Transactions on Computers
Y distributed application platform
Computer Communications
Computer Performance Modeling Handbook
Computer Performance Modeling Handbook
Federating Traders: An ODP Adventure
Proceedings of the IFIP TC6/WG6.4 International Workshop on Open Distributed Processing
Load balancing in homogeneous broadcast distributed systems
Proceedings of the Computer Network Performance Symposium
Load Sharing in Distributed Systems
IEEE Transactions on Computers
Hi-index | 0.24 |
We consider the case of trading in an environment of autonomous components. Trading is the process of matching a service request with the offers to support that particular service made by potential servers. In the case of multiple matching offers a selection of the most suitable server can be made according to the properties of the server. Among the possible selection policies we concentrate on those taking into account the actual load of the server. The selection process can be delegated to components of a special type, called traders. Various aspects influence the performance of the server selection process, including the delay of the load information, the selection strategy, the use or non-use of traders, and the number of concurrent traders, if any. We investigate several trading strategies in a scenario where the load information is broadcast periodically, and where multiple traders may operate concurrently.