The Raritan Blog

Raritan IT publishes measured power savings by Power IQ

Allen Yang
October 22, 2010

In late August 2010, I wrote on this blog last time that Raritan IT will use Power IQ and PX to monitor the incremental power consumption of the Cisco UCS, while we gradually migrate some physical rack-mount physical servers as VM onto UCS blades.  The following results reflect our P2V migration activities between 2010-08-05 and 2010-10-20.  We have 2 UCS chassis that we call UCS1 and UCS2, and both of them are installed on our Cisco Rack 4.  During the time period from 08/05 to 10/20, we had the following physical activities:

  • 08/11: Turn on two blades on UCS for Engineering testing
  • 08/15: Turn on 2nd 10Gib port on Fabric Interconnect
  • 08/30: Remove 2 VMware ESX Servers (the 2 Dell PowerEdge 1950)
  • 09/01: Turn 1 additional blade on UCS2 to measure its effect on power consumption
  • 09/03: Remove SharePoint Test Server (a Dell PowerEdge 2850) and DNS2 (a 1U PC-server)
  • 09/14: Migrate BES as a VM onto UCS blade (an HP Proliant DL140)
  • 09/20: Migrate Exchange Server (a Dell PE1950) as a VM onto UCS blade; turn testing blade off
  • 10/11: Turn one UCS blade on for Oracle JDE remote DR site

From Power IQ data reading we have the following observations:

  1. Each UCS blade we have consumes roughly 4.3 KWh per day.   This is somewhat consistently demonstrated from measuring the total power consumption of the observed systems including the 2 UCS chassis and the non-UCS physical servers.  On 08/11, the measurement was 334.6 KWh.  On 08/12, 24-hours later after turning on 2 UCS blades, the measurement was 339.7KWh.  Since then the UCS power consumption stayed at that level until the next even on 08/15.  This 4.3KWh incremental power consumption per UCS blade per day was also consistently measured between 09/01 and 09/02 when IT turned on the 3rd blade on UCS2 to measure incremental power consumption.
  2. Migrating mildly loaded physical rack-mount servers as VMs onto the UCS blades typically doesn’t increase much power consumption.  We Installed VMware ESX host on UCS and migrated two physical application servers onto UCS blade, and we couldn’t even find material difference before and after adding these 2 VMs onto UCS.  But the power savings from decommissioning physical servers can be clearly seen.  On 08/30, we removed 2 ESX host physical servers (Dell PowerEdge 1950), the total power consumption dropped by 6.55KWh a day.  Then on 09/03, we removed another PowerEdge 2850 (a Sharepoint test Server) and a 1-U PC server (DNS); and we saw a drop of power consumption by 7.2KWh a day.   Therefore, we see a clear power saving advantage from migrating physical servers into VMs.
  3. The new version of Power IQ is very helpful in data exporting and giving power consumption insights at various levels of details.  We captured several diagrams from Power IQ to see the trend over time, then right from there we export the data into Excel spreadsheet, and we can examine the day by day details in there.

In summary, between 08/05 and 10/20, Raritan IT measured power saving of 20KWh a day with the activities of migrating 5 physical servers as VMs onto Cisco UCS blades and remove these physical servers from the racks.  The daily power consumption of these observed systems combined dropped from slightly over 330KWh a day on 08/05 to slightly under 310 KWh a day on 10/20.  The Power IQ trend diagram also portrays such gradual reduction, which very accurately reflects the activities conducted and the resulting effects on power consumption over the observed time period. 

Power Consumption from Aug 5, 2010 to Oct 20, 2010

Raritan IT will continue its physical server migration activities in the next few months, there may be more interesting findings we can gather in the future; stay tuned.

(part of this blog entry is also posted in response to a comment for my late August blog post)


Raritan will be at the Green Data Center Conference in NYC on October 20th and 21st

Dorothy Ochs
October 11, 2010

Raritan will be at the Green Data Center Conference on Oct. 20-21, 2010 at the Embassy Suites New York (102 North End Avenue, New York, NY).  Come check us out to learn about the latest in power management solutions.


Why CIOs Should See Their Power Bills

Susana Thompson
September 30, 2010

Data center infrastructures have grown and evolved in step with technology advancements.  In  amassing  the most-advanced technologies and powerful IT equipment to process the ever-growing data spawned by the Internet era, data centers today are sprawling information fortresses  —reaching  over one million square feet and processing terabytes and petabytes of information daily.  The creation of the modern data center focused on uptime and availability, and without much thought to energy— so it is no surprise that today’s data centers are not energy efficient.  With their  insatiable appetite for energy and skyrocketing energy costs, data centers face a number of power-related problems.  


Raritan is co-presenting "400V Data Center Solutions" with Starline at the Sept. 30th DatacenterDynamics show in Chicago.

Dorothy Ochs
September 15, 2010

We will be displaying our intelligent PDU’s, KVM-over-IP switch, and DCIM solutions in our booth. Stop by and visit us - you’ll be entered into our drawing for a Canon digital camcorder.  For more information and to register: https://www.datacenterdynamics.com/en/conferences/.


Are you over-cooling your Data Center?

James Cerwinski
September 14, 2010

You should consider the following steps if think you are over-cooling your data center or lab — this will likely result in significant energy savings.

1.  Start monitoring — Deploy temperature and humidity sensors according to industry recommendations(the top, middle and bottom of the server inlet side of the rack).  You can conveniently connect sensors to rack PDUs.

2.  Determine if you are over-cooling — Deploy a system to chart your readings to indicate if you are operating within the ASHRAE 2008 recommended range.  In 2008, ASHRAE increased their recommended upper temperature limit to 80.6F.

3.  If you are over-cooling consider gradually increasing your temperature.

4.  Monitor for hot spots — Use a system that will generate events and notifications when non-critical and critical temperature and humidity thresholds are violated.

5.  Be aware of long term trends — Chart maximum and minimum environmental readings for long term trend analysis.

 

James Cerwinski


Page 78 of 92 pages ‹ First  < 76 77 78 79 80 >  Last ›