Adaptive parallelism with Piranha
Adaptive parallelism with Piranha
Interfacing Condor and PVM to harness the cycles of workstation clusters
Future Generation Computer Systems - Special issue: resource management in distributed systems
Medusa: an experiment in distributed operating system structure
Communications of the ACM
The EASY - LoadLeveler API Project
IPPS '96 Proceedings of the Workshop on Job Scheduling Strategies for Parallel Processing
Building and Scheduling Parallel Adaptive Applications in Heterogeneous Environments
IWCC '99 Proceedings of the 1st IEEE Computer Society International Workshop on Cluster Computing
An Efficient Resource Allocation Scheme for Gang Scheduling
IWCC '99 Proceedings of the 1st IEEE Computer Society International Workshop on Cluster Computing
| Hi-index | 0.00 |
In this paper, we present a framework for exploiting resources in Networks of Workstations (NOWs) and Clusters of Processors (COPs), in order to run multiple parallel adaptive applications. Adaptive model includes parallel applications capable to adapt their parallelism degree dynamically following availability of resources and changes in the underlying environment's state. Within the framework of the proposed environment, many components are developed. This includes fault tolerance, adaptive application building and scheduling and multiapplication scheduling. In this paper, we focus our study on the multiapplication scheduling problem. In the proposed multi-application scheduling model, each parallel adaptive application is controlled by its own scheduler, responsible for optimizing resources used by the application. A dynamic multi-application scheduler supervises all the applications and shares resources fairly among them, by means of a combined (time-sharing and space-sharing) scheduling algorithm.