People/Web Search Calendar Emergency Info A-Z Index UVA Email University of Virginia

Computer Science Colloquia

Tuesday, May 27, 2014
In Kee Kim
Advisor: Marty Humphrey
Attending Faculty: Marty Humphrey (Chair), Kevin Skadron, Kevin Sullivan, and Yanjun Qi

2:00 PM in Rice 242

Ph.D. Qualifying Exam Presentation
Comprehensive Elastic Resource Management to Ensure Predictable Performance for Scientific Applications on Public IaaS Clouds

Scientists have become increasingly reliant on large scale compute resources on public IaaS clouds to efficiently process their applications. Unfortunately, the reactive nature of auto-scaling techniques made available by the public cloud provider can cause insufficient response time and poor job deadline satisfaction rates. To solve these problems, we designed an end-to-end elastic resource management system for scientific applications on public IaaS clouds. This system employs the following strategies: 1) an accurate and dynamic job execution time predictor, 2) a new resource evaluation scheme that balances cost and performance, and 3) an "availability-aware" job scheduling algorithm. This comprehensive system is deployed on Amazon Web Services and is compared with other state-of-the art resource management schemes. Experimental results show that our system achieves a 28%-73% improvement with respect to the deadline satisfaction rate over other schemes. We achieve this deadline satisfaction rate improvement while still providing improved cost-efficiency over other state-of-the-art approaches