Standards can be a good thing, especially in IT (think USB, Firewire etc). Standards for measurement are great too, such as the Metric system which is used by most of the known world. Sorry couldn't resist that one America…
What is more relevant and also very imporant to us folks here at 1E is a standard for measuring Data Center Efficiency, which is just one of the things we're about. Fortuitously enough then, that news reaches me today via the global telegraph , that good sense has prevailed in the world of Data Center metrics in the form of an agreed global standard in this field. Almost.
The Green Grid has announced that agreement has been reached with various agencies including the U.S. Department of Energy’s Save Now and Federal Energy Management Programs, U.S. Environmental Protection Agency’s ENERGY STAR Program, European Commission JRC Code of Conduct, Japan’s Ministry of Economy, Trade and Industry, Japan’s Green IT Promotion Council, and The Green Grid.
These participants have agreed on the use of Power Usage Effectiveness (PUE) as the preferred energy efficiency metric. PUE as a measurement metric has been around for a while but hopefully this agreement will see the metric go global.
This is important so that Data Centers of all shapes, sizes and nationality can be measured with the same set of rules, and it will also help to quantify the effectiveness of so called 'Green IT' initiatives with a recognised standard. In plain English, that means the CFO's may be a little happier about paying for it!
One important entry in the agreement, which I feel has been a tad unreported, is in the following paragraph of the document.
Effective energy efficiency metrics that:
1. Measure the actual IT work output of the data center compared to actual energy consumption. It is of note that in the process to define IT work output, the following interim measurements are being defined and / or validated:
a. IT – Measure the potential IT work output compared to expected energy consumption; and measure operational utilization of IT equipment
b. Data center facility and infrastructure – Measure the data center infrastructure efficiency (PUE)
So, as you can see I've highlighted the bit that excites me – just in case you missed it. This is what we've been trying to get over to people here at 1E for years, and that is the fact that it's all very well measuring the amount of energy that comes into the data center, and you can even measure things at the individual rack or UPS level, but to achieve true efficiency you need to monitor individual servers to see just what they are doing. Just because a server is on, well, it don't mean a thing.
In the next paragraph of the agreement, it states:
The collective groups are in agreement on the following guiding principles, as an interim step toward the desired outcomes (1. b.). It is recommended that data centers begin to measure PUE according to these principles:
- Power Usage Effectiveness (PUE) using source energy is the preferred energy efficiency metric. PUE is a measurement of the total energy of the data center divided by the IT energy consumption
- The industry should improve the IT measurement capabilities to ultimately enable taking the measurement directly at the IT load (e.g. servers). At a minimum IT energy measurements should be measured at the output of the UPS
- For a dedicated data center, total energy measurement should include all energy sources at the point of utility handoff. For data centers in larger buildings, total energy should include all cooling, lighting, and support infrastructure, in addition to IT load
So what they are saying here, is that in order to make your efficiency measurement as accurate as possible, you need to be monitoring your servers. Which is exactly why we introduced the concept of Useful Work™.
Using NightWatchman Server Edition you can monitor the exact metrics that this agreement is talking about, and integrate the results into your overall efficiency stats. Our monitoring capability in NWSE tells you exactly what a server is doing and when. For instance it can highlight when backups are running (Useful Work™), or when a rogue application is chewing up the CPU and/or disk (Really Not Useful Work). Having identified a potential slacker server you can then implement our Drowsy Mode which is where the real savings start. BTW Really Not Useful Work is something I just made up but I quite like it and may submit it to the Green Grid for the next draft – what do you think?
In summary, I think that this agreement signifies a great leap forward in Data Center efficiency measurement. China and India have yet to get involved but I'm sure it's just a matter of time, and then we will at least have a chance to implement true global energy savings.
Useful Work Whitepaper
NightWatchman Server Edition Homepage
The Agreement in full (Green Grid)