data center cooling

Aug 6 2009   8:28PM GMT

Future Facilities updates data center CFD tool

Posted by: Matt Stansberry
DataCenter, cfd, data center cooling

London-based Future Facilities released Version 5 of its data center computational fluid dynamics software, 6SigmaDC. Large data centers and facility design consultants (including Emerson/Liebert’s data center consulting arm) use the tool to model data center cooling.

New features in version five include:

-Modeling internally cooled cabinets so they can be registered for inventory purposes and accounted for in the power system as well as for power scaling needs.
-Modeling equipment weight to assess whether any given layout will breach floor loading limitations.
-Cable penetrations can now be attached to the raised floor to account for the situation where they are installed before racks and cabinets.
-The infrastructure outside the facility can now be included in graphical views and animation can be generated to demonstrate how airflows develop.

While data center pros rave about the tool (6SigmaDC won silver in our 2007 Products of the Year) it comes at a steep price, up to three times the cost of data center CFD software from TileFlow, according to data center design consultant Pete Sacco.

Jul 6 2009   4:28PM GMT

Emerson energy exec calls for data center water metric

Posted by: Mark Fontecchio
data center cooling

Jack Pouchet, the director of energy initiatives for Emerson Network Power, is calling for a new metric measuring a data center’s water use as compared to its productivity (measured in Emerson’s own Compute Units per Second).

In a column in Environmental Leader, Pouchet writes that water could be “the next oil,” meaning that the availability of water is often difficult for many in the world, especially clean drinking water in developing nations. Pouchet suggests that the data center industry should be more cognizant of the water it’s using to cool IT equipment, and adding a water use metric to the current Power Usage Effectiveness (PUE) metric — which compares facility energy use to IT energy use — is a good start.

From Pouchet’s column:

It is time for the data center industry to formulate a Water Systems Productivity metric (WSP). Take useful work or even a proxy for useful work, such as the proposed Compute Units Per Second, and divide that by the amount of water used during the period. Water may be measured in units, with 1 unit equal to an acre-foot. However, gallons/liters is also acceptable.

This WSP metric would ideally be reported monthly with your other metrics. Once we start to measure and report water utilization, we will quickly realize that simply flowing more cooling water in order to “economize” may not always be the best answer. Now we will be able to have a meaningful tool to determine the ideal mix between dry-coolers, CW plants and evaporative cooling towers compared to the increased energy used with alternative solutions.

Jul 2 2009   8:27PM GMT

Equinix wins energy savings award from Silicon Valley group

Posted by: Mark Fontecchio
data center cooling, Airside economizers, Variable frequency drive (VFD) fans

Major data center colocation company Equinix scored an energy award from the Silicon Valley Leadership Group this week for the energy-efficient design its data centers.

Equinix has received almost $1 million in rebates from the city of Santa Clara for the energy efficiency measures, which include using airside economizers and variable frequency drive (VFD) fans in their data center infrastructure equipment.

The company estimates that the airside economizers save $300,000 per year and the variable speed fans save $51,000 per year. See “Data center air-conditioning fans blow cost savings your way” and “Data center cooling: airside and waterside economizers” for more information on these cost-saving data center technologies.

Jun 30 2009   3:16PM GMT

Yahoo! to reportedly turn data center 10 degrees for better cooling

Posted by: Mark Fontecchio
data center cooling

The designs for a new Yahoo! data center in western New York have changed - by 10 degrees.

The engineering firm for the future 180,000-square-foot facility told local planning officials in Lockport, N.Y., that it was changing its design plans by 10 degrees to improve cooling. This according to Orest P. Ciolko of the Wendel Duchscherer engineering firm.

“That has to do with the prevailing winds during the months we need cooling,” Ciolko told The Buffalo News. Ciolko said that computer modeling done by Yahoo!’s designers “realigned the pods so the prevailing winds will blow directly into louvers on the sides of the buildings,” according to the story.

May 21 2009   6:40PM GMT

Building data centers in Afghanistan

Posted by: Mark Fontecchio
data center cooling, Container Data Center

A couple weeks ago I got the chance to spend the morning with Paul Brenner, who works in the high-performance computing department at the University of Notre Dame. Brenner is spearheading a project to build a containerized data center next to a local municipal greenhouse so that, during winter months, the heat from the servers can be piped into the greenhouse to warm it up. Check out the Notre Dame greenhouse data center story (there’s a cool video).

Another thing I learned from Brenner when hanging out with him is that he is an engineering officer in the U.S. Air Force Reserves, and actually just returned from an overseas deployment in Afghanistan a few weeks ago. While there, Brenner helped build data centers.

Obviously it’s not an ideal place, and Brenner had to do a lot of improvising. A few things complicated his mission. First, with it being the military, so much information is siloed, with select people able to access it. So not only do different branches of the military want their own data centers and their own servers, but divisions within each branch want close control of their IT assets. So a lot of data centers there are hodgepodge, small, and consist of a rack here or a rack there.

Brenner mentioned how some of the major IT vendors out there, such as IBM and HP and Sun Microsystems, have tried pitching their containerized data centers as a suitable option for military operations. But Brenner said that even in ideal conditions, deployment time is measured in months. In many cases, Brenner doesn’t have that much time.

So he made do. Oftentimes he would take a bunch of household air-conditioning units and daisy-chain them together, which he said actually led to a good deal of cooling redundancy. It’s all about adjusting to conditions, and when your overseas serving your country in a barren desert land, you do whatever you can to keep the computers running.

Feb 23 2009   8:03PM GMT

Data center high density vs. low density: Is there a paradox?

Posted by: Mark Fontecchio
Data center power, data center cooling

Over at CIO, there is an article about a so-called data center power-density paradox. According to Michael Bullock, the CEO of a data center design consultancy called Transitional Data Services, if you don’t beware the power-density paradox, “it will ensnare you in an unappetizing manner.”

OK, so what is it? Bullock argues that as you increase the power density in your data center, “your efforts to free up space in your data center could boomerang, creating an even greater space crisis than you had before.”

Drilling down, the paradox says that as you use more dense equipment (which places greater demands on power and cooling), you will quickly reach an inversion point where more floor space is consumed by support systems than is available to your IT equipment – typically between 100 and 150 watts per square foot. This translates into greater capital and operational costs, not the reductions you were hoping to achieve.

How much space will you need?  At a power density of about 400 watts per square foot, plan to allocate about six times your usable data center space for cooling and power infrastructure.  So before you embrace high-density as a quick fix to your space problem, make sure you have adequate room to house the additional power and cooling infrastructure, sufficient raised-floor space to handle the increased airflow demands of hotter-running boxes and, of course, sufficient available power to operate the hungry systems and their support gear. If any of these resources are unavailable or inadequate, your data center will not support the increased power density. And you will have wasted your time and money.

Let’s drill down, though, for real. Let’s say you decide your data center needs to process more. As an example, let’s say you need to expand your data center so that you have 1,024 processing cores, which you calculate as 256 quad-core processors. Should you use a power-dense design, such as blade servers, or spread that processing power out amongst 1U rack servers?

Hewlett-Packard’s c7000 BladeSystem enclosure is your blade server design. In a 42U rack, you can fit four c7000s, each of which can hold 16 HP ProLiant BL2×220c G5 server blades, for a total of 64 quad-core Intel Xeon 5400 processors. That adds up to 256 quad-core processors in a single rack. Each c7000 chassis demands 6 x 2,400 watts of power, or 14,400 watts. Multiply that by four chassis and you have 57.6 kilowatts of power in a single rack holding 256 quad-core chips.

Now, let’s use a spread-out design with HP’s DL100 rack servers. A DL160 G5 rack server is 1U and holds one quad-core Intel Xeon 5400 processor. So it will take 256U, or about six 42U racks, to reach the same processing power as a single BladeSystem. Each DL160 server demands 650 watts of power, so 256 of them demands 166.4 kilowatts of power.

To sum up:

• Power-dense design: 1,024 processing cores using blade servers use up 42U of space and 57.6 kilowatts of power
• Less power-dense design: 1,024 processing cores using 1U rack servers use up 256U of space and 166.4 kilowatts of power

According to this, there is no power-density paradox. If you use power-dense equipment, you will use less space and less power.

Now, I realize that cooling a single rack of blade servers would be a ridiculously difficult chore, and would take a lot more effort than a single rack of rack servers. But that would be comparing a single blade server rack to six racks full of rack servers. It’s not an apples-to-apples comparison.

Bullock’s point is not lost. If you have a rack of 1U servers, don’t expect to be able to convert that rack to blade servers and provide the same level of power and cooling infrastructure as you presently have. It won’t happen. But that’s a comparison of more processing power to less processing power. Comparing equivalent processing power designs yield no paradox, at least on the power side of the equation.

The cooling side of the equation is a different story, and can be complicated by factors such as airside economizers, which can cool less-dense data centers but can’t cool a 57.6KW rack. So as an example, if you spread your IT equipment out enough, then maybe you could eliminate mechanical chillers altogether. That could not only cut down on space, but on cost as well (which is what matters in the end). Also, your raised floor might be able to cool six racks of 1U servers with normal CRAC units, but you might need to convert to overhead or liquid cooling to cool a single 57.6KW rack properly.

In any event, the issue is not as simple as Bullock makes it out to be. Power-dense equipment will not always lead to more data center power and cooling equipment. Oftentimes, it will lead to less when matched up against a comparable rack-server design.

Feb 2 2009   7:04PM GMT

Data center fan efficiency hubbub from ASHRAE

Posted by: Mark Fontecchio
Data Center airflow, data center cooling

The following is a guest post from Vali Sorell, an associate partner in the critical facilities division at the Syska Hennessy Group. Sorell was a speaker at the ASHRAE Technical Committee (TC) 9.9 sessions in Chicago last week, and had some comments about our story on data center air-conditioning fans.

---

We all agree that reducing the fan speed saves energy, regardless of whether that fan is driven through a VFD (variable frequency drive) or if that fan is driven by an EC (electronically commutated) motor. The person from 365 Main stated that “it’s not worth the extra cost to have fans run at a lower speed for such a short time.” That misses the point - the fact is that it IS possible to save energy at all times, and that lower speed does not occur for a short time. That lower speed occurs forever! This is best explained by an example.

Assume a data center has 100 CRAC units, 80 of them are needed to meet the loads, and 20 of them are needed for redundancy. This amounts to 25% redundancy, which is very typical for most data centers. Let’s also assume that the load is constant from now till forever (meaning that the part load conditions are not considered, i.e. they are history).

Case 1: 80 units running at 100% speed consume 80/100 = 80% of the possible fan energy use.
Case 2: 100 units running at 80% speed consume 80/100 x 80/100 x 80/100 = 51% of the possible fan energy use.

Compared to normal operation, using all of the available redundant CRAC units at variable speed (regardless of whether that variable speed is achieved by EC motors or VFD’s) consumes 100%-(51%/80%) = 36% less fan energy than running the load-required complement of CRAC units at full speed. That’s not a small amount of energy!

Jan 28 2009   5:39PM GMT

More on ASHRAE’s expanded recommended data center temperature, humidity ranges

Posted by: Mark Fontecchio
data center cooling

In a story earlier this week, I reported on the American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) TC 9.9 expanding its recommended guidelines for data center temperature and humidity ranges. One person who wanted that range to expand beyond what it did was Christian Belady, the principal power and cooling architect at Microsoft. Belady gave me the following image to make his point (click the picture for a much larger version):

This is a typical psychrometric chart that looks at dry bulb temperature, wet bulb temperature, dew point, relative humidity and other factors. Because there are several variables involved, you often get these odd shapes when plotting ranges.

As the key in the upper left shows, the solid red block was ASHRAE’s previous recommended range. The red outline is its current recommended range. The blue block is ASHRAE’s allowable range, which ASHRAE defines as an environment able to support IT equipment running, but with marked breaking down of equipment compared to the recommended range.

Then comes the yellow block, which is the typical vendor specification. Belady’s argument is that ASHRAE should be pushing beyond that yellow block, not cozying up within it. If the vendors will warranty equipment within that yellow block, why should ASHRAE have a range inside it?

Jan 27 2009   10:31PM GMT

Perforated tiles can be your friend — even in the hot aisle

Posted by: Mark Fontecchio
data center cooling, Data Center airflow

With the American Society of Heating, Refrigerating and Air-conditioning Engineers expanding its recommended data center temperature range, you have to wonder: Should you start wearing a bathing suit when doing data center maintenance?

ASHRAE TC 9.9 now recommends data center temperatures as high as about 80 degrees Fahrenheit. But that is to be measured at the location of the server inlet. How about on the other side, in the hot aisle? The difference between cold and hot aisles, often referred to as Delta T or just ΔT, can be as much as 50 degrees Fahrenheit. Which means hot aisle temperatures could approach 130 degrees Fahrenheit, and if the equipment is live, that means 130-degree air blowing in your face. Not exactly ideal working conditions.

So what to do, ASHRAE members wondered? Some think that server manufacturers need to start redesigning their boxes so they can be accessed and maintained from the front in addition to, or instead of, from the back. That way data center staff could work in the more-tolerable cold aisle where heat stroke is less likely.

Another option is simply to pull up a tile where you’re going to work in the hot aisle, and replace it with a perforated tile. That way you can get a nice chilly gust of cold air blasting from your feet to counteract the furnace blowing in your face. Sure, putting perforated tiles in the hot aisle is considered a severe no-no in well-designed hot/cold aisle data center configurations.

But if it’s temporary, and it can prevent the need to have an IV bag of fluids on site just in case of severe dehydration and overheating of employees, well, then it might be worth it.

Oh, and if you’re not working in a raised-floor environment, you might be out of luck. Maybe you can invest in a couple oscillating fans.

Jan 26 2009   9:37PM GMT

Raise the data center temperature, but bring your earplugs

Posted by: Mark Fontecchio
data center cooling, Data Center airflow

CHICAGO - The American Society for Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) has expanded recommended guidelines for ambient data center temperatures, but they’re warning that the new range could cause data center noise increases.

There has been a lot of talk recently about raising data center temperatures to improve energy efficiency, as the air conditioners don’t have to work as hard to cool the room. ASHRAE TC 9.9 recently changed its recommended upper data center temperature from 77 degrees Fahrenheit (25 degrees Celsius) to 80.6 degrees Fahrenheit (27 degrees Celsius). Munther Salim, a mechanical engineer at HP EYP Mission Critical Facilities, said raising the set points in CRAC units is the “number one thing you can do to save money.”

Google is raising data center temperatures. So is Microsoft and Intel. But Michael Patterson, a thermal engineer at Intel, warned that raising the data center temperature could have an effect on “acoustical noise levels.”

“Servers with (variable frequency drive) fans on servers — the increase in power comes mostly from the increase in fan power after 25 degrees Celsius,” he said. “Servers in 27 degrees Celsius may have higher acoustics due to higher fan speed.”

Patterson showed the following graph:

As you can see, fan power (the orange-reddish line) rises exponentially after about 25 degrees Celsius, which happens because of the increase in fan speed to keep the server components cool enough in a warmer environment. According to an ASHRAE document on the extended environmental envelope, “it is not unreasonable to expect to see increases in the range of 3-5″ decibels if the ambient temperature increases from 25 to 27 degrees Celsius.

“Data center managers and owners should therefore weigh the trade-offs between the potential energy efficiencies with the proposed new operating environment and the potential increases in noise levels,” the document states.

Are there any other solutions? Some suggest reversing the trend of server manufacturers to miniaturize components. Just a few years ago, 1U servers might have one single-core processor. Now they might have multiple quad-core chips. That leads toward having to dissipate more heat in a smaller space. If servers were made bigger, the fans wouldn’t have to work as hard to do so much in such a tight area.

But that might not be feasible for some users. According to a SearchDataCenter.com survey earlier this year, 32% of users say that lack of space is most limiting their data center growth. Making servers bigger won’t help that.