Autonomic Power Management Schemes for Internet Servers and Data Centers

Lykomidis Mastroleon; Nicholas Bambos; Christos Kozyrakis; Dimitris Economou

doi:10.1109/GLOCOM.2005.1577776

Autonomic Power Management Schemes for Internet Servers and Data Centers

Lykomidis Mastroleon

Nicholas Bambos

Christos Kozyrakis Stanford

Dimitris Economou

IEEE Global Telecommunications Conference (GLOBECOM), 2005

DOI: 10.1109/GLOCOM.2005.1577776

Abstract

We investigate autonomic power-control policies for Internet servers and datacenters. By monitoring system load and thermal status, the policies vary the processing resources in use to balance delay and power. A dynamic-programming formulation captures this tradeoff, while low-complexity heuristics achieve performance close to the optimal policy and improve on standard benchmarks.

Materials

Bibtex

@inproceedings{2005autonomicpowermanagementschemesforinternetserversanddatacenters,
  title = {Autonomic Power Management Schemes for Internet Servers and Data Centers},
  author = {Lykomidis Mastroleon and Nicholas Bambos and Christos Kozyrakis and Dimitris Economou},
  booktitle = {IEEE Global Telecommunications Conference (GLOBECOM)},
  year = {2005},
  doi = {10.1109/GLOCOM.2005.1577776},
}

Hi, we are the Stanford MAST team

Autonomic Power Management Schemes for Internet Servers and Data Centers

Abstract

Materials

Bibtex