ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 A Novel Market Oriented Dynamic Collaborative Cloud Service Infrastructure
Cited 12 time in scopus Download 0 time Share share facebook twitter linkedin kakaostory
저자
Mohammad Mehedi Hassan, Biao Song, 윤장우, 이현우, 허의남
발행일
200909
출처
World Conference on Services 2009, pp.9-16
DOI
https://dx.doi.org/10.1109/SERVICES-2.2009.20
협약과제
09MR5700, IPTV 융합서비스 및 콘텐츠 공유를 위한 개방형 IPTV 플랫폼 기술개발, 류원
초록
In this paper, we present a novel combinatorial auction (CA) based Cloud market model that facilitates dynamic collaboration (DC) among Cloud providers (CPs) for providing composite/collaborative Cloud services to consumers and hence can address the interoperability and scalability issues for Cloud computing. Also to minimize the conflicts that may happen when negotiating among providers in a DC platform, we propose a new auction policy in CA that allows a CP to dynamically collaborate with suitable partner CPs to form a group before joining the auction and to publish their group bids as a single bid to fulfill the service requirements completely. But to find a good combination of CP partners is a NP-hard problem. So we propose a promising multi-objective (MO) optimization model for CP partner selection that not only uses their individual information (INI) but also their past collaborative relationship information (PRI) which is seldom considered in existing approaches. A multiobjective genetic algorithm (MOGA) called MOGA-IC is also developed to solve the model. We implemented our proposed CACM model and the MOGA-IC in a simulated environment and study their economic efficiency and performance with existing model and algorithm. The experimental results show that the proposed MOGA-IC can support satisfactory and high quality partner selection in CACM model. © 2009 IEEE.
키워드
Cloud market, Combinatorial auction, Dynamic collaboration, MOGA, Partner selection
KSP 제안 키워드
Cloud Computing, Cloud market, Cloud providers, Combinatorial Auction, Efficiency and Performance, Existing Approaches, Market model, Multi-objective Genetic Algorithm, NP-Hard problem, Optimization Model, Partner selection