Storage @ Work

Robin Harris, the StorageMojo, has a couple interesting posts on his ZDNet blog and his StorageMojo blog.

If you are interested in this topic, I think both are worth a look see.

Robin Harris at StorageMojo has an excellent post on energy consumption of systems. Worth a read.

Definitely one of the coolest things about being with Dell is the push for all things green. The Regeneration.org site is a terrific resource that I'm hoping to tap into both professionally and personally.

It was great to see Fujitsu join The Green Grid. Way to Go!.

There are plenty of challenges ahead to make data centers greener. The US Environmental Protection Agency (EPA) has taken a large interest in the energy efficiency of Internet data centers. With an expected huge surge in cloud computing, it's very smart to start figuring out ways to keep our information infrastructure from consuming more energy then necessary.

Computerworld had a news piece today on their plans to help create comparison metrics for data centers so they can be awarded the Energy Star seal.

Vendors are doing things too. In my previous post, I welcomed Fujitsu to the Green Grid.

Here's a link to a video with Dell engineering manager Tom Garvens, who discusses the energy saving designs of Dell's M-series blade chassis products. Products like this would be a great way to control energy consumption.

The Environmental Protection Agency (EPA) wants to get in front of the energy consumption estimates for data centers in the next several years and find a way to rate technology products and data centers for their energy efficiency.

I attended a meeting today about the future of energy standards for storage products. This is fairly serious work and it involves the DMTF, Green Grid and SNIA - three groups that are working to identify the measurements and methods that will eventually lead up to an EPA Energy Star label for storage products.

It's a messy process to be sure, but there are a lot of smart people working on this. The challenge is coming up with measurements that can be applied across a large set of products and configurations and representing an amazingly diverse set of workloads. A zillion decisions need to be made about what gets measured, how it gets measured, how that data is stored and and compiled and eventually interpreted. The three industry groups involved are now trying to figure out how to divide the work so they can be as efficient as possible.

From an operating perspective, it looks like there will be a lot more monitoring of system operations and health in the future. It sort of reminds me of the transformation we've seen in cars and trucks over the last 50 years. You can look under the hood of a 1965 pickup truck and see all the belts, hoses, sparkplugs, wires and everything else you might want to inspect and change. If you look under the hood in a new car today, its hard to see what's what with all the hoses and things that manage the efficiency and output of the engines.

Tomorrow's servers and storage won't have hoses, belts and tubes shoved in their cases, but they will use a fair amount of processing power to monitor and manage their environmentals. Processing power not used for data processing, but for running as efficiently as possible.

This post first appeared on Inside IT, a Dell blog.

HP got an early jump on April Fool’s day yesterday when they announced their new blade server, the Proliant BL260c G5, with the claim that it has 64% better energy efficiency than the Dell PowerEdge M600 blade server. The details of this Internet Vaudeville act can be found here: http://h71028.www7.hp.com/enterprise/downloads/HP_ProLiant_BL260_SPECjbb2005_032808a.pdf

Normally, vendors try to establish comparisons using similar configurations, such as those done by Principled Technologies here:

http://www.principledtechnologies.com/Clients/Reports/Dell/Win2K8_blades0208.pdf

Yes, Dell commissioned this work, but if you look at Appendix B of that report, you'll notice that a professional rigor was applied to making sure there was a level playing field. By contrast, the HP report published yesterday makes no attempt at comparing similar configurations. Different processors, running in different power modes with different memory and disk configurations, etc.

If HP had substituted one of their own servers in place of the M600 – one that was actually comparable to the M600, such as the BladeSystem c-Class tested by Principled Technologies, the power savings would probably have been over 75%, instead of the 64% they reported. But they couldn't very well come out and say that their high mileage lab experiment was so much better than an HP product that many of their customers already own.

The Server Room has to be one of the best forums I've seen. So if Dell EqualLogic people go there, lets make sure we don't trash it. Put the Name Dell in your sign on to avoid being poachers and don't pimp our stuff. It's fine to talk about experiences and what we know and all that, but its a tech forum not a drop box for hype - perbole.

And the rest of you too. Do your thing and back up your blood type. Somebody might need it for the ultimate restore.