TA2197

Distributed Power Management: Technical Deep Dive

VMware Distributed Power Management (DPM) saves power in an ESX server
cluster by consolidating virtual machines onto fewer hosts and powering
off hosts during periods of low resource utilization, and powering hosts
back on for virtual machine use when workload demands increase. DPM is
an optional add-on to VMware Distributed Resource Scheduling.
This talk will explain how DPM operates, and present some example
scenarios in which use of DPM can save significant power in a DRS cluster,
while operating in concert with DRS and (optionally) HA constraints.

Technical Overview Session

Networking; Storage / Backup; VMware Infrastructure; Virtualization Platform / Hypervisor

1 Hour

Anthony Vecchiolla( Int'l.Integrated Solutions ), Canturk Isci( VMware, Inc. ), Anne Holler( VMware, Inc. )

Anthony Vecchiolla is the lead VMware Solutions Architect for International Integrated Solutions based in Plainview, NY. He has nearly 8 years experience with VMware products including large scale Virtual Infrastructure implementations within the Pharmaceutical and Financial sectors.

Canturk Isci is a performance engineer in VMware, focused primarily on distributed resource management performance. Canturk has a PhD in electrical engineering from Princeton University, NJ, an MSc in VLSI system design from University of Westminster, UK, and a BS in electrical engineering from Bilkent Universiy, Turkey.

Anne Holler has worked at VMware since January 2002 as an engineer,
previously in the Performance Team and now in the Distributed Resource
Management Team. She previously worked in the compiler low-level optimizer
team at Hewlett-Packard and in Transmeta's high-performance/low-power group.
She holds a doctorate in computer science from University of Virginia and
B.S. and M.S. degrees in computer science from Duke University.