Software to Keep Servers Running During Cooling Failures

Purdue has created software for Linux that will slow down processors during a cooling failure in a data center.

While a processor runs, it generates heat. The slower it runs, the less heat it generates. Thus, when the air cooling system in a data center fails, the less heat the better. When thousands of servers are clocked downwards, the heat savings will be tremendous.

With the software from Purdue, a server will slow way down in order to generate the least amount of heat possible. With this change, servers can actually be kept running longer and thus could potentially avoid downtime entirely.

At Purdue’s supercomputing center where this was developed, they’ve already survived several cooling failures without downtime.

Purdue’s situation, however, does appear to have some unique qualities. One is that the software was designed for their clusters, which number in the 1,000s of CPUs – meaning that activating a slow-down can happen across several thousand servers simultaneously. This has a tremendous affect on the cooling in the data center and also becomes easy since all the servers are identical.

With that many servers, the cluster can dominate the server room as well. In a heterogenous environment like most corporate server rooms, software like this would have to be on all platforms to be effective.

The places that slowdown software could be most effective is in large clustered environments, as well as small or homogenous environments. Slowdowns could be triggered by many things: cooling failures, human intervention, or even heating up of the server itself.

Current Ethernet Not Enough?

At the recent Ethernet Technology Summit, there was grousing going on about the need for more power-conservative switches, for more manageable switches, but most of all for faster Ethernet.

Facebook, for one, spoke of having 40Gbits coming out of each rack in the data center, and related how its 10Gb Ethernet fabric is not enough and won’t scale. There are new standards (100Gb Ethernet and Terabit Ethernet) but they are not yet finalized. Analysts suggest that there is a pent-up demand for 100Gb Ethernet, and the conference bore that out.

Supposedly, there is supp

Energy Star Program for Data Centers

The EPA announced that they are expanding the Energy Star Program to include data centers; the measurements are expected to be finalized in June 2010.

The EPA is hoping that the new Energy Star rating for data centers will become a selling point for data centers. The new rating is based largely (but not completely) on the PUE (or Power Usage Effectiveness). William Kosik wrote an article in the September 2007 issue of Engineered Systems Magazine that explains PUE quite well and in detail.

Google talks about their efforts for power-efficient computing in their data centers in some depth; it’s very interesting.

IBM also announced just recently that they are building a new data center in Research Triangle Park where they will test effect of various temperature levels in the data center – and will cool it with outside air as well.

This is definitely an exciting time for data center power research; seems that there is something new every day.

Data Centers: Weta Digital, New Zealand

Weta Digital, the special effects company behind Lord of the Rings, King Kong (2005), X-Men, and Avatar is in the news again.

Data Center Knowledge has an article about their data center, as well as another one about it last year.

Information Management also had an article about it, as well as a blog post by Jim Ericson.

HP even has a video about their use of HP blades in their cluster.

Some of the more interesting things about their data center include:

  • The use of water-cooling throughout.
  • Using external heat exchangers to release heat.
  • Using blades in a clustered configuration.

This is just the beginning. While this data center is not as radical as the others discussed here recently, the data center is more in the realm of current possibilities. There are photographs in the current Data Center Knowledge article as well.

An Experimental Underground Data Center: Iron Mountain’s Room 48

Iron Mountain has converted an old mine in Pennsylvania to a computing facility, part of which includes an experimental energy-efficient data center that uses geothermal conditions to improve cooling.

ComputerWorld wrote an article about their tour of the facility, including Room 48 where the experimental data center is housed. The power distribution transformers and the air conditioning units are outside the data center, rather than inside it. It also relies on the extreme pressure differential between the hot and cold aisle to move the air, making the data center very quiet (in contrast to the usual data center).

The data center in Room 48 operates at between 70 and 72 degrees Fahrenheit, and operates 200 watts per square foot (more than 50% above the usual 125 watts per square foot).

This data center is truly remarkable.

A Data Center in a Silo

The CLUMEQ project is designing a supercomputer, and has several sites already built. One of these, a site in Quebec, was built in an old silo that used to contain a van de Graaf generator.

An article in the McGill Reporter from several years ago described the supercomputer installation at Montreal.

The new CLUMEQ Collossus (as the Quebec installation is called) was described in an article in Data Center Knowledge. The design has all of the computers (Sun blades) are in a circle with the core being a “hot core” and the cool air being drawn from the rim.

The Green500 List

The Green 500 ListFor many years now, there has been a list of the Top 500 supercomputers in the world as measured by computational speed. But is speed the only metric that a supercomputer should be measured by?

Almost certainly not. In recent years, the discussion has changed as awareness of data center energy consumption has become an important topic. These top supercomputers are, in many cases, egregious offenders of energy consumption.

Thus, the Green 500 list was born: a list of supercomputers (taken from the Top 500) arranged by energy efficiency. This makes for interesting reading, and helps raise the awareness that computational power is not the only reason to buy a computer.