SGER--CSR/SMA: Thermal Aware Dynamic Resource Management for Datacenters

点击翻译按钮获取中文摘要

Minimizing the energy cost and improving thermalperformance of datacenters are among the key issues towards optimizing computingresources and maximally exploiting a datacenter's computation capability. In general,there is no automated feedback control loop between the physical environment of a datacenter, and the datacenter's management software/system software scheduler for makingthermal and power-efficient resource management decisions. The energy cost and thermaldistribution of a datacenter depends on the multi-scale properties of a large number ofmulti-modality system variables, such as cooling capacity, hardware thermalcharacteristics, and workload profiles. An integrated, system-oriented approach towardsthis problem could provide several significant benefits.Thiswork, will investigate approachers for a unique merger between physical infrastructure and the resourcemanagement functions of the cluster operating system to take a holistic view of data centermanagement, and make global (at the level of a datacenter), power-aware, andthermal-aware job scheduling decisions. The project seeks to develop a comprehensive solutionfor on-line, dynamic and automated thermal management and control framework for thepower-limited datacenters. The methods include an integrated two-pronged approach for incorporating fast thermal evaluation, using an abstract heat model, in the resource managementfunctions of the datacenters; namely: new thermal aware scheduling techniques will be developedby making traditional scheduling techniques cognizant of thermal performance of theirscheduling decisions by taking into account the spatio-temporal thermal implications ofjob placements. This entails making scheduling decisions by taking into account not onlyjob characteristics but also physical placement of computational node, the thermalcharacteristics of heterogeneous systems, and thermodynamics. These thermal-awarescheduling techniques will be comprehensively evaluated using new performance metricssuch as utilization per operating cost.