skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Measuring Server Energy Proportionality

 [1];  [1]
  1. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
Work for Others (WFO)
OSTI Identifier:
DOE Contract Number:
Resource Type:
Resource Relation:
Conference: The 6th ACM/SPEC International Conference on Performance Engineering (ICPE), Austin, TX, USA, 20150131, 20150204
Country of Publication:
United States

Citation Formats

Hsu, Chung-Hsing, and Poole, Stephen W. Measuring Server Energy Proportionality. United States: N. p., 2015. Web.
Hsu, Chung-Hsing, & Poole, Stephen W. Measuring Server Energy Proportionality. United States.
Hsu, Chung-Hsing, and Poole, Stephen W. 2015. "Measuring Server Energy Proportionality". United States. doi:.
title = {Measuring Server Energy Proportionality},
author = {Hsu, Chung-Hsing and Poole, Stephen W},
abstractNote = {},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 2015,
month = 1

Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Energy consumption in datacenters has recently become a major concern due to the rising operational costs andscalability issues. Recent solutions to this problem propose the principle of energy proportionality, i.e., the amount of energy consumedby the server nodes must be proportional to the amount of work performed. For data parallelism and fault tolerancepurposes, most common file systems used in MapReduce-type clusters maintain a set of replicas for each data block. A coveringset is a group of nodes that together contain at least one replica of the data blocks needed for performing computing tasks. In thiswork, we develop and analyze algorithmsmore » to maintain energy proportionality by discovering a covering set that minimizesenergy consumption while placing the remaining nodes in lowpower standby mode. Our algorithms can also discover coveringsets in heterogeneous computing environments. In order to allow more data parallelism, we generalize our algorithms so that itcan discover k-covering sets, i.e., a set of nodes that contain at least k replicas of the data blocks. Our experimental results showthat we can achieve substantial energy saving without significant performance loss in diverse cluster configurations and workingenvironments.« less
  • HEPIC an information {open_quotes}center of centers{close_quotes} for the HEP community, is a 24 hour online location where a HEP researcher can start her/his search for information. Operated by the HEP Network Research Center, HEPIC is accessible via WWW, gopher, anonymous FTP, DECnet, and AFS. This paper describes HEPIC`s design and future plans, and the HEPNRC`s efforts to collect information and link high energy physics researchers world-wide.
  • Fifty-seven percent of US servers are housed in server closets, server rooms, and localized data centers, in what are commonly referred to as small server rooms, which comprise 99percent of all server spaces in the US. While many mid-tier and enterprise-class data centers are owned by large corporations that consider energy efficiency a goal to minimize business operating costs, small server rooms typically are not similarly motivated. They are characterized by decentralized ownership and management and come in many configurations, which creates a unique set of efficiency challenges. To develop energy efficiency strategies for these spaces, we surveyed 30 smallmore » server rooms across eight institutions, and selected four of them for detailed assessments. The four rooms had Power Usage Effectiveness (PUE) values ranging from 1.5 to 2.1. Energy saving opportunities ranged from no- to low-cost measures such as raising cooling set points and better airflow management, to more involved but cost-effective measures including server consolidation and virtualization, and dedicated cooling with economizers. We found that inefficiencies mainly resulted from organizational rather than technical issues. Because of the inherent space and resource limitations, the most effective measure is to operate servers through energy-efficient cloud-based services or well-managed larger data centers, rather than server rooms. Backup power requirement, and IT and cooling efficiency should be evaluated to minimize energy waste in the server space. Utility programs are instrumental in raising awareness and spreading technical knowledge on server operation, and the implementation of energy efficiency measures in small server rooms.« less