Grouping and deploying fine-grained tasks on grid by learning performance data
When deciding the size or the granularity of a batch, one should consider the utilisation constraints imposed on the resources by their respective providers; e.g. the maximum time allowed for task execution and the maximum allowed storage space. In addition, the size of the batch should not overload...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2011
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | When deciding the size or the granularity of a batch, one should consider the utilisation constraints imposed on the resources by their respective providers; e.g. the maximum time allowed for task execution and the maximum allowed storage space. In addition, the size of the batch should not overload the interconnecting network. The main objective of this thesis is to study the factors involved in deciding a batch size and design the relevant batch resizing policies and techniques. The policies and techniques are then developed and experimented in a small-scale grid environment. Throughout the conduct of this thesis, the batch resizing policies and techniques were aligned accordingly to support various purposes which led to several following major findings and contributions |
|---|