作者: Zahra Abbasi , Georgios Varsamopoulos , Sandeep K. S. Gupta
关键词: Distributed computing 、 Provisioning 、 Efficient energy use 、 Data center 、 Heuristics 、 Computer science 、 Server 、 Computer network 、 The Internet 、 Cloud computing 、 Server farm
摘要: With the increasing popularity of Internet-based information retrieval and cloud computing, saving energy in Internet data centers (a.k.a. hosting centers, server farms) is importance. Current research approaches are based on dynamically adjusting active set order to turn off a portion servers save without compromising quality service; workload then distributed, conventionally equally (i.e. balanced), across servers. Although there ample work that demonstrates savings through dynamic provisioning, little thermal-aware provisioning. This paper provides formulation thermal aware provisioning (TASP), nonlinear minimax binary integer programming form, series heuristic solving them, namely MiniMax, bb-sLRH, CP-sLRH sLRH. Furthermore, it introduces distribution (TAWD) among The proposed heuristics evaluated using model ASU HPCI center, while request traffic real web traces 1998 FIFA World Cup as well SPECweb2009 suite. TASP found outperform power-aware-only selection scheme (CPSP), by up 9.3% for simulated scenario. achieved efficiency is: MiniMax (9.3% savings), (9.2%), bb-sLRH (8.6%), sLRH (5.8%), compared CPSP.