|
Web
Education
Viewing 1-10 of 10 total results
Achieving Energy Proportionality In Server Clusters
a great amount of interests in the past few years. Energy proportionality is a principal to ensure that energy consumption is proportional to the system workload. Energy proportional design can effectively improve energy efficiency of computing systems. In this paper, an energy proportional model is proposed based on queuing theory and service differentiation in server clusters, which can ......
Achieving Energy Proportionality In Server Clusters
CiteSeerX - Document Details (Isaac Councill, Lee Giles, Pradeep Teregowda): Green computing is a hot issue that has received a great amount of interests in the past few years. Energy proportionality is a principal to ensure that energy consumption is proportional to the system workload. Energy proportional design can effectively improve energy efficiency of computing systems....
Achieving Energy Proportionality In Server Clusters - CORE
Achieving Energy Proportionality In Server Clusters . By Xinying Zheng and Yu Cai. Abstract. a great amount of interests in the past few years. Energy proportionality is a principal to ensure that energy consumption is proportional to the system workload. Energy proportional design can effectively improve energy efficiency of computing systems ......
EPSCS: Simulating and Measuring Energy Proportionality of ...
Energy proportionality for a server cluster means energy consumption is proportional to the workloads running on the cluster. One problem is that it is too costly, time-consuming, and complex to build a real cluster to evaluate energy-proportional algorithms. Aiming to solve this problem, we propose to build a prototype system that is able to ......
 An Experimental Evaluation of Datacenter Workloads On Low ...
on the approach of achieving energy e ciency. The rst category is to seek energy proportionality with non-energy-proportional servers [53, 32, 31, 27, 35]. The second cate-gory is to build more energy-e cient architecture based on low-power CPU [38, 41, 49, 33, 43, 46, 29]. Improving energy-proportionality allows power draw to...
Reducing Cluster Energy Consumption through Workload ...
Energy consumption is a major and costly problem in data centers. For many workloads, a large fraction of energy goes to powering idle machines that are not doing any useful work. There are two causes of this inefficiency: low server utilization and a lack of power proportionality. We focus on addressing this problem for two workloads: (1) a traditional, front-end web server workload and (2 ......
WU ET AL.: ON ENERGY SECURITY OF SERVER SYSTEMS 865 On ...
2.2 Energy Proportionality Energy proportional computing [4] is an important concept in today’s server systems. It aims to address the increasing energy concern and demand for power saving by making servers consume energy proportional to its workload. This goal is normally achieved by conditionally trading off performance for power savings....
Energy Proportionality and Workload Consolidation for ...
processors [22, 45] provide a foundation for energy propor-tionality. The second goal is workload consolidation, which raises server utilization and minimizes the number of servers needed for a particular set of workloads [14, 46, 50]. Ad-vances in cluster management [11, 32] and server consoli-dation using virtual machines or container systems ......
Reducing Cluster Energy Consumption through Workload ...
Energy consumption is a major and costly problem in data centers. For many workloads, a large fraction of energy goes to powering idle machines that are not doing any useful work. There are two causes of this inefficiency: low server utilization and a lack of power proportionality.
Energy proportional datacenter networks
tween common server workload profiles and server energy effi-ciency [3]. In particular, they show that a typical Google cluster spends most of its time within the 10-50% CPU utilization range, but that servers are inefficient at these levels. They therefore make the call for energy proportional computing systems that ideally con-...
|
