HP’s PUE article, quotes HP Employee Chris Malone, Ahh Chris works for Google Now

October 5, 2008 Dave Ohara

Found this article on HP’s Enterprise Magazine, discussing PUE for data center efficiency. At the time it was written, Chris Malone was an HP employee, but now he works for Google. Transforming Your Enterprise Magazine

Winter 2008

Metrics drive efficiency in the data center

In an effort to understand where data center energy inefficiencies lie, companies are adopting standardized metrics of PUE and CPE to determine the initiatives required for optimization.

Energy consumption and inefficiency within data centers is a well publicized problem and a key concern for IT organizations. Gartner has predicted that, "By 2011, more than 70 percent of U.S. enterprise data centers will face tangible disruptions related to floor space, energy consumption and/or costs."*

Before data center inefficiency can be resolved, experts say it must be characterized and tracked in a consistent manner.

“Our investigations have shown that most data centers are inefficient, expending more power for cooling than for running the IT equipment,” notes Dr. Chris Malone, Thermal Technologies Architect for HP. “Until recently, there was no established method for benchmarking data center energy consumption, making it extremely difficult to characterize the full extent of the problem.”

Malone suggests a standardized evaluation of and rating for data center energy consumption to help companies determine:

How well is a data center running relative to others in the industry?

Are new data centers more efficient than the old ones?

Which data centers should be retired?

Do best practices really improve efficiency?

Do new technologies actually perform as suppliers promise?

HP didn’t quote Christian Belady who was a PUE innovator at HP.

For those of you who didn’t know Christian is now at Microsoft. Chris is at Google. Too bad HP couldn’t keep these guys at HP. Otherwise, we would be discussing HP’s innovations in data center efficiency.

Compare Sun’s PUE/Data Center Disclosure vs. Google’s

October 5, 2008 Dave Ohara

Sun’s PR team sent me some blog entries on their PUE efforts.

One is Sun discussing their achieving a 1.28 PUE.

I'll show you mine...

So how efficient is your datacenter?

Last month I received some pretty cool news. The Chill-Off we have been hosting for the Silicon Valley Leadership Group (SVLG) in their datacenter demonstration project, was completed. This was a head to head test against APC In-Row, Liebert XDV, Spraycool, Rittal Liquid Racks and IBM Rear Door Heat Exchanger. Sun was the host that provided the datacenter, plant, compute equipment, and support. Lawrence Berkeley National Labs (LBNL) was the group conducting the test on behalf of the California Energy Commission (CEC) The results from this test will be published in a report in June of this year and we will be hosting the event.

But one piece of information came out of this that I could not wait to share. As part of the chill-off, LBNL did a baseline of our plant in Santa Clara. Mike Ryan on my staff then captured the usage data for the remaining portions of the datacenter. This gave us our PUE or DCiE (pick you favorite) number.

Surprise, Surprise!

I knew that our datacenter would be efficient because of the way we had designed it, but I did not have any data to back it up yet. We had been telling our customers, and others that toured the center, that we are pretty sure we would be under the industry targeted PUE of 2 (Created by Christian Belady from the Green Grid). That was a conservative number. But when I got the data back, even I was surprised at how efficient it was.

We achieved a PUE of 1.28!

Google says they eliminated data for facilities below 5MW.

Such a strong claim demands evidence, especially in light of recent criticism of companies "gaming the numbers." On this page we will explain our measurements in detail to ensure that they are realistic and accurate. It is worth noting that we only show data for facilities with an actual IT load above 5MW, to eliminate any inaccuracies that can occur when measuring small values. This section is aimed at data center experts, but we have tried to make it accessible to a general technical audience as well.

But given Google’s claim of measurement accuracy, it would seem like they should be able to measure numbers below 1MW like Sun has.

Error Analysis

To ensure our PUE calculations are accurate, we performed an uncertainty analysis using the root sum of the squares (RSS) method. Our uncertainty analysis shows that the overall uncertainty in the PUE calculations is less than 2% (99.7% confidence interval). Our power meters are highly accurate (ANSI C12.20 0.2 compliant) so that measurement errors have a negligible impact on overall PUE uncertainty. The contribution to the overall uncertainty for each term described above is outlined in the table below.

Here are more technical details on Sun’s PUE calculations.

Sun exposes more details than Google as they want your business.

Better yet, would you like Sun to help design your datacenter to achieve the same efficiencies? Let us drive the next generations physical and technical solutions for you. Sun's Eco Virtualization practice can do just that. Email me at dean.nelson@sun.com and I'll tell you how.

What is Google’s motivation for its PUE disclosure?

Google’s PUE Coverage

October 3, 2008 Dave Ohara

It has been interesting watching Google’s PUE and data center claims spread through blogs and news.

I haven’t blogged about Google’s post, even when I saw it on other blog sites.

Why? Because, I knew Google was going to write a PUE post, and by the time they finally did, I was disappointed it didn’t provide more facts. Nothing really new, other than Google claims being the most efficient.

As ComputerWorld mentions, the release of the PUE #’s coinciding with another marketing event.

This announcement coincides with a speech about energy that CEO Eric Schmidt gave Wednesday evening in San Francisco, in which he proposed a $4.4 trillion clean-energy plan.

It is great that Google has helped to educate thousands of people on PUE and how much energy can be saved.

There are some details, but not enough to help on equipment selection.

I plan on writing more on this and contrasting what others have shared on PUE best practices.

Electricity Home Monitoring Solutions, 19 ZigBee products registered

September 29, 2008 Dave Ohara

Ready for the confusion, 19 home electricity monitoring solutions have been registered.

In recent months, 19 energy-management products have been certified for ZigBee interoperability. These include communicating thermostats costing $50 to $200, controllers costing $10 to $60, and in-home displays costing $50 to $200.

The WSJ reports on the momentum for home electricity monitoring solutions for the home.

People Power

The more you know about your energy use, the less energy you may use

A revolution in smart electronics is going to give consumers unprecedented ability to control energy consumption and save money.

Making it possible is the introduction of inexpensive electronics -- including thermostats and controllers -- that can be assembled into simple home networks able to control equipment such as air conditioners, lights, water heaters and furnaces.

The home systems, in turn, are the building blocks of a smart grid that seamlessly flows information on energy usage and cost from consumers to utilities and back again.

There are lots of vendors out there.

In August, ZigBee joined forces with the HomePlug Powerline Alliance, which promotes devices that use a building's wiring as the communications path. So wireless ZigBee devices soon will be able to work with HomePlug devices, which are being developed by companies such as Cisco Systems Inc., Comcast Corp., Intel Corp. and Texas Instruments Inc.

What happens with consumers behavior is the unknown.

In Boulder, Xcel Energy Inc. is ramping up a program, through its Public Service of Colorado utility, where customers will test the company's technologies and report back about what works and what doesn't. The goal is to get about 10,000 metered customers, about 20% of the city's population, doing some sort of energy management in the next year or so.

"We want to be somewhat agnostic and not influence people, because we're trying to learn more ourselves," says Sandy Simon, director of utility innovations and smartgrid strategy for Xcel.

Many test homes will be equipped with technology from GridPoint Inc. that connects to an information portal operated by the Arlington, Va., firm. GridPoint Chief Executive Peter Corsell says consumers will log onto a site and home networks will be able to tell them how much they can save, in dollars or kilowatt hours.

"You'd say, 'Cut my bill,' and it will find a way that's least painful for you," says Mr. Corsell. The system "might interrupt your water heater from 11 p.m. to 5 a.m., for example, or change the setting on your thermostat."

Sun's Intelligent Power Monitoring Service

September 27, 2008 Dave Ohara

Sun has announced a new download for Sun HW and Solaris OS to monitor the power consumption.

Sun Intelligent Power Monitoring Service

Sun's Intelligent Power Monitoring Service is a Smart Energy solution for the data center that makes it possible for you to improve power efficiencies by monitoring, reporting and forecasting energy consumption, cooling requirements and load utilization of server and storage devices.
Start Now

Key Benefits

Make more efficient use of energy consumption, cooling requirements and load utilization for server and storage devices in your data center while maintaining system uptime, so you can:

More effectively plan for capacity expansion or unexpected changes by accurately assessing and forecasting dynamic power consumption.
Reduce your carbon footprint by ensuring adherence to eco-responsible policies.
Know whether your eco-efficiency investments return real energy savings.
Lower energy costs and take advantage of energy rebates that yield additional savings.

Key Applications

This affordable service is ideal for all data centers: small, medium and large. Provides immediate insight to help you:

Reduce data center operating costs by optimizing energy consumption
Enable departmental charge-back for energy consumption
More accurately plan for facility expansions, new sites or moves
Better manage power utility or facility constraints
Meet emerging regulatory requirements for energy reporting
Take advantage of financial incentives from energy companies

I am going to try the folks at Digital Sense as they use Sun blade servers for their high density computing, and se if they have given this solution a try.