TY - GEN
T1 - Allocation of discrete energy on a cloud-computing datacenter using a digital power grid
AU - Xu, Yifei
AU - Rojas-Cessa, Roberto
AU - Grebel, Haim
PY - 2012
Y1 - 2012
N2 - Electrical power supply of datacenters constitutes major operational costs. In this paper, we focus on providing the supplying energy to exact match the consumption demands by using the properties of a digital power grid. Different from the present power grid, where energy is supplied continuously and in amounts to satisfy large demands, the digital power grid supplies energy in the form of packets and upon request. Energy packets are specifically addressed to one, or multiple servers and to no others, and they carry the amount of energy that has been requested. To evaluate the level of energy supply, we model the consumption time of a server as an ON-OFF Markov chain. Through simulation, we show that the digital grid not only reduces the energy safety margins used by current energy providers to assure system stability but it also highly satisfies the energy requests of datacenter servers when the capacity of the power grid equals the average energy request and without a safety margin.
AB - Electrical power supply of datacenters constitutes major operational costs. In this paper, we focus on providing the supplying energy to exact match the consumption demands by using the properties of a digital power grid. Different from the present power grid, where energy is supplied continuously and in amounts to satisfy large demands, the digital power grid supplies energy in the form of packets and upon request. Energy packets are specifically addressed to one, or multiple servers and to no others, and they carry the amount of energy that has been requested. To evaluate the level of energy supply, we model the consumption time of a server as an ON-OFF Markov chain. Through simulation, we show that the digital grid not only reduces the energy safety margins used by current energy providers to assure system stability but it also highly satisfies the energy requests of datacenter servers when the capacity of the power grid equals the average energy request and without a safety margin.
KW - Carbon footprint
KW - Cloud computing
KW - Datacenters
KW - Green cloud
KW - Resource sharing
KW - Task scheduling
UR - http://www.scopus.com/inward/record.url?scp=84875492017&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84875492017&partnerID=8YFLogxK
U2 - 10.1109/GreenCom.2012.94
DO - 10.1109/GreenCom.2012.94
M3 - Conference contribution
AN - SCOPUS:84875492017
SN - 9780769548654
T3 - Proceedings - 2012 IEEE Int. Conf. on Green Computing and Communications, GreenCom 2012, Conf. on Internet of Things, iThings 2012 and Conf. on Cyber, Physical and Social Computing, CPSCom 2012
SP - 615
EP - 618
BT - Proceedings - 2012 IEEE Int. Conf. on Green Computing and Communications, GreenCom 2012, Conf. on Internet of Things, iThings 2012 and Conf. on Cyber, Physical and Social Computing, CPSCom 2012
T2 - 2012 IEEE International Conference on Green Computing and Communications, GreenCom 2012, 2012 IEEE International Conference on Internet of Things, iThings 2012 and 5th IEEE International Conference on Cyber, Physical and Social Computing, CPSCom 2012
Y2 - 20 November 2012 through 23 November 2012
ER -