Research Article
MSIS: Meta-scheduler Infornation Service for Resource Performance Predictors
@INPROCEEDINGS{10.1109/COMSWA.2007.382447, author={Afrasyab Bashir and Atulya K. Nagar and Hissam Tawfik}, title={MSIS: Meta-scheduler Infornation Service for Resource Performance Predictors}, proceedings={2nd International IEEE Conference on Communication System Software and Middleware}, publisher={IEEE}, proceedings_a={COMSWARE}, year={2007}, month={7}, keywords={Availability Computerized monitoring Condition monitoring Delay Distributed computing Intelligent systems Laboratories Optimal scheduling Resource management State estimation}, doi={10.1109/COMSWA.2007.382447} }- Afrasyab Bashir
Atulya K. Nagar
Hissam Tawfik
Year: 2007
MSIS: Meta-scheduler Infornation Service for Resource Performance Predictors
COMSWARE
IEEE
DOI: 10.1109/COMSWA.2007.382447
Abstract
Contemporary Grid monitoring services lack the functionality to monitor batch-schedulers' job queue information and the status of jobs submitted. Moreover, the monitored data delivered by them is only virtually real-time. The former implies that these services are not capable of monitoring a batch-scheduler's inability to process a particular job. The latter implies that the strategies based solely upon the currency of monitored information can lead to delayed, if not wrong, scheduling and management decisions. Batch-scheduler monitoring is fundamentally imperative for future resource performance predictions that help meta-schedulers in deciding optimal schedules for deadline-constrained jobs. We have developed a meta-scheduler information service (MSIS), which monitors the batch-schedulers for up-to-date information about their job queues, current status of the jobs and the underlying resources. MSIS can be used in its two totally independent modes: as a retrofit API library to enhance capabilities of contemporary grid monitoring services and as a web service. It, in combination with any grid monitoring service, provides comprehensive information which can be used in estimating a resource's future state over a limited period of time.