Resource selection techniques in large-scale distributed systems for dynamic task sharing

다국어 초록 (Multilingual Abstract)

Large-scale distributed system provides an attractive scalable infrastructure for network applications. In such environment there exist large sets of heterogeneous and geographically distributed resources. Among numerous optional resources, selecting appropriate resources for applications is challenging and affected by many factors. The loosely coupled nature of large-scale distributed environment makes data access unpredictable and instability. Slow allocation process may offset the benefit obtained by running a job on a fast node. Besides, the operation condition of a resource provider changes rapidly. The status of job execution and computing capability of a resource provider need to be considered dynamically. In this paper we present an approach of dynamic task-sharing based on the record of previous data download and current execution status of resource providers to select the appropriate one or more providers to execute a job together. The proposed approach can also avoid single point failure and server overload problem.

목차 (Table of Contents)

• Abstract 6
• 1. Introduction 8
• 2. The Related Works 11
• 2.1 Resource selection 11
• 2.2 Data accessibility 14
• 3. The Proposed Scheme 17
• 3.1 System model 17
• 3.2 Performance measurement criteria 18
• 3.3 Resource selection policy 25
• 4. Performance Evaluation 32
• 4.1 Experiment setup 33
• 4.2 Evaluation result 34
• 4.1.1 Comparison of the elapsed time between existing and proposed scheme 34
• 4.1.2 The effect about the different unit of providers 35
• 5. Conclusion 38
• References 39
• Korean Abstract 42