Set Up DPM and Use vCenter Scheduled Task to Turn DPM Off and On

August 19, 2011 in vHersey, VMware

VMware Power Management or Distributed Power Management (DPM) powers down hosts when resources are not needed and powers the host on as the demand for resources increases. This feature has the potential to provide significant savings in power and cooling especially in a large cluster.

I recently implemented DPM on the VMware Lab at TCC, a 12 host cluster that is only utilized a couple nights a week for classes. When hosts are idle DPM powers them down (Standby Mode) and then powers them up during class times as student access VMs requiring additional cluster resources. With DPM enabled there are usually only 2 or 3 of the 12 hosts powered on outside of class hours when there is very little resource consumption . This greatly reduces the load on the UPS and the heat in the TCC server closet room.

After successfully getting DPM working in the school lab and a bit of testing I decided to set it up on my production environment. One difference between the school lab and the production environment (besides the size – 4 host instead of 12) is that I want ensure all of my host resources are available during business hours. I do not want to have to wait for hosts to power on when resources are needed especially at the start of the business day. This can be accomplished by configuring DPM and then setting up a vCenter Scheduled task to turn DPM off and on.
Read the rest of this entry →

Awesome HA and DRS Audit Script Based on the HA/DRS Technical Deepdive

May 17, 2011 in VMware

If you do not have a copy of VMware vSphere 4.1 HA and DRS Technical Deepdive by @DuncanYB and @FrankDenneman then go here, buy it, and read it cover to cover. The book covers everything VMware High Availability (HA) and the Distributed Resource Scheduler (DRS) (and then some!).

Alan Renouf, co-author of another must have book – VMware vSphere PowerCLI Reference: Automating vSphere Administration, has put together a nice HA and DRS Audit PowerCLI script that can be used to gather information and check for best practices on your HA/DRS cluster based on the information from the Technical Deepdive book.

The script is run against your vCenter and creates a nicely formatted HTML report with all the HA/DRS related details of each cluster managed by vCenter. This gives you a detailed overview of your environment and the configuration of HA and DRS.

As of right now the script is based on the information contained in just the first 50 pages of the HA and DRS Deepdive. Alan Renouf is promising updates to the script as he works his way through the book.

Go here to download the HA and DRS Audit script.