Thesis
A self-managed cloud cache for accelerating data intensive applications
Washington State University
Master of Science (MS), Washington State University
2012
Handle:
https://hdl.handle.net/2376/101403
Abstract
Certain classes of data-intensive applications, e.g., data analysis and scientific workflows, are known to contain redundant overlaps in their computational pattern indicating that these applications can benefit from caching intermediate and final computed results for reuse. We propose a fully autonomous and easily deployable cloud cache with the objective of accelerating compute- and data-intensive processes. Our system, which is distributed over multiple cloud nodes, intelligently provisions resources at runtime based on user’s cost and performance expectations, while abstracting the various low-level decisions regarding efficient cloud resource management and data placement within the cloud from the user. The crux of our system is a mathematical prediction model employing Multi-Objective Optimization that engages an Artificial Neural Network to gain insight into future workload that might soon commence onto the system. The prediction model lends the system the capability to auto-configure the optimal resource requirement to automatically scale itself up/down, i.e., allocate or deallocate cloud nodes, to accommodate demand surge/lulls while staying within user’s cost constraint as well as optimizing user’s cost and performance expectations. Our multi-tiered cloud cache is implemented and evaluated utilizing various Amazon Web Services (AWS) Infrastructure as a Service (IaaS) features
Metrics
2 File views/ downloads
18 Record Views
Details
- Title
- A self-managed cloud cache for accelerating data intensive applications
- Creators
- Farhana Mannan Kabir
- Contributors
- David T. Chiu (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; Pullman, Wash. :
- Identifiers
- 99900525037601842
- Language
- English
- Resource Type
- Thesis