Constrained Component Deployment in Wide-Area Networks Using AI Planning Techniques
IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Optimal Resource-Aware Deployment Planning for Component-Based Distributed Applications
HPDC '04 Proceedings of the 13th IEEE International Symposium on High Performance Distributed Computing
Analysis of deployment dependencies in software components
Proceedings of the 2006 ACM symposium on Applied computing
Dependability in Software Component Deployment
DEPCOS-RELCOMEX '07 Proceedings of the 2nd International Conference on Dependability of Computer Systems
A planning method for component placement in smart item environments using heuristic search
DAIS'07 Proceedings of the 7th IFIP WG 6.1 international conference on Distributed applications and interoperable systems
Improving availability in large, distributed component-based systems via redeployment
CD'05 Proceedings of the Third international working conference on Component Deployment
A decentralized redeployment algorithm for improving the availability of distributed systems
CD'05 Proceedings of the Third international working conference on Component Deployment
| Hi-index | 0.00 |
Internet of machines also known as Machine-To-Machine is a system composed of thousands of interconnected machines that inter operate to run an application that meets a specific need. Such an application can be implemented by the encapsulation and the cooperation of several software entities called component. These components will be distributed on nodes with different resources in terms of memory, capacity of calculation, bandwidth, battery dependent, etc. The placement of these components on the nodes is a crucial step for both the application and the target machines. The choice of placement actually influences their performance. Such a deployment algorithm has to take into account the functional needs, constraints and properties as well as the available resources and performance of the whole system. Many algorithms have been proposed for this, including one based on a principle of auction, DecAp [13], which allows the dynamic redeployment of components in order to take into account the evolution of the system's resources. We present in this article an extension of this algorithm by adding the management of available resources on each node: memory, bandwidth and computing capacity.