Presentation 2010/7/28
Development of middleware for data processing load distribution using cloud computing resource
Shiori TOYOSHIMA, Saneyasu YAMAGUCHI, Masato OGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, data management and IT cost have become serious problem as the advanced IT society progresses. Cloud computing is introduced in order to save the construction cost of local cluster at each organization. This mechanism is suitable for supplying insufficient CPU power to a computing system. However, data-intensive applications have a totally different feature from CPU-intensive applications. Thus, execution of data-intensive applications making use of cloud computing resources is discussed in this paper. Since a data processing system with better cost/performance ratio is required, we have constructed middleware for load distribution among cloud computing resources and a local cluster. Scalable resource management is achieved by monitoring resource usage of a local cluster and insufficient resources are acquired dynamically from a cloud service. Our middleware allocates injected jobs optimally among the local cluster and commercial cloud service Amazon EC2, based on the result of the I/O disk load conditions. According to the evaluation using a data-intensive benchmark, our middleware has achieved an excellent job allocation, which is close to an ideal case in terms of the cost of the cloud service.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # CPSY2010-23
Date of Issue

Conference Information
Committee CPSY
Conference Date 2010/7/28(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Computer Systems (CPSY)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Development of middleware for data processing load distribution using cloud computing resource
Sub Title (in English)
Keyword(1)
1st Author's Name Shiori TOYOSHIMA
1st Author's Affiliation Ochanomizu Univ()
2nd Author's Name Saneyasu YAMAGUCHI
2nd Author's Affiliation Kogakuin University
3rd Author's Name Masato OGUCHI
3rd Author's Affiliation Ochanomizu Univ
Date 2010/7/28
Paper # CPSY2010-23
Volume (vol) vol.110
Number (no) 167
Page pp.pp.-
#Pages 6
Date of Issue