Scheduling tight deadlines for scientific workflows in the cloud
Cloud computing has increasingly become a demand for scientific computations as it provides users with simple access for computation. Commercial clouds are also used for scientific analysis and computation because of their scalability, latest high-quality hardware as well as pay-per-use cost model....
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2018
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/69027/1/FSKTM%202018%2054%20-%20IR.pdf http://psasir.upm.edu.my/id/eprint/69027/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Cloud computing has increasingly become a demand for scientific computations as it provides users with simple access for computation. Commercial clouds are also used for scientific analysis and computation because of their scalability, latest high-quality hardware as well as pay-per-use cost model. Commercial clouds can be easily accessed globally. There have been several studies presenting new algorithms to generate deadline constrained schedules to minimize the execution cost as well as the high failure rate in schedule constructions. However, there are increased failure rates whenever tight deadlines are produced.
The work in this paper focuses on the hurdle of scheduling tight deadline scientific workload. This article will evaluate the performance of the Proportional Deadline Constrained (PDC) algorithm using Cloudsim and compare it with the Deadline Constrained Critical Path (DCCP) scheduling algorithm. The performance evaluation is done using two different performance metrics, success rate and normalized cost. The results show that the PDC performs better in term of success rate metric while the DCCP algorithm has better performance in term of normalized cost metric. The PDC could be improved on the normalized cost. |
|---|