Capacity Planning in Economic Grid Markets

Risch, Marcel; Altmann, Jörn

doi:10.1007/978-3-642-01671-4_1

Cited by 3 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an economic grid, the main priority is given to the application budget and deadline [2,5]; one would process as many tasks as possible within the given budget and deadline. The authors in [28,30] introduced capacity planning when executing application jobs in a commercial or an economic grid. The term capacity covers the deadline, budget, storage space, resource utilisation, and load balancing.…”

Section: Related Workmentioning

confidence: 99%

“…The resource providers impose charges on the users for consuming the CPU cycles or storage space of their resources. Apart from gaining profit, this helps to control the usage of the resources [30,28,10,26]. In such an environment, we propose to optimally map the tasks to the resources based on the QoS which we refer to as advance QoS planning: the task-resource mapping is performed prior to the task group deployment in order to maximise the processed task count within the QoS.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

QoS-based Task Group Deployment on Grid by Learning the Performance Data

et al. 2014

View full text Add to dashboard Cite

Overhead of executing fine-grain tasks on computational grids led to task group or batch deployment in which a batch is resized according to the characteristics of the tasks, designated resource, and the interconnecting network. An economic grid demands an application to be processed within the given budget and deadline, referred to as the quality of service (QoS) requirements. In this paper, we increase the task success rate in an economic grid by optimally mapping the tasks to the resources prior to the batch deployment. The task-resource mapping (Advance QoS Planning) is decided based on QoS requirement and by mining the historical performance data of the application tasks using a genetic algorithm. The mapping is then used to assist in creating the task groups. Practical experiments are conducted to validate the proposed method and suggestions are given to implement our method in a cloud environment as well as to process real-time tasks. Response to Reviewers:Reviewer #2:(1) Authors should better discuss complexity issues of their proposal, even at a theoretical level.Answer: This has been addressed briefly in section 7.2.4 Issues and Future Direction.As shown in Fig. 5, the proposed batch resizing strategy requires the meta-scheduler to keep monitoring, learning and updating the average deployment metrics for each task category-resource pair using a genetic algorithm. A frequent con-duct of advance QoS planning may delay the entire task group deployment as the genetic algorithm will increase the computation overhead at the meta-scheduler. With the current high-end machines, the overhead or latency can be reduced or hid-den by configuring the genetic algorithm to keep running in parallel as a separate thread. The meta-scheduler can obtain the latest, optimal task-resource mapping at any time from the genetic algorithm.

show abstract