The Evolution of Data Center Efficiency Metrics

In April 2006, Christian Belady (now Director of Green Grid and Principal Infrastructure Architect Microsoft Global Foundation Services) gave one of the most important symposium presentations in Data Center history at the Uptime Institute Forum. It was the first time the concept of PUE had been showcased, and from that the Green Grid was formed and the PUE metric became a reality.

This is an important milestone in history as it was the first time ever that there was an independent Data Center power and performance metric that was relatively easy to understand, implement and measure.

So what exactly is PUE – In simple terms, PUE is the total facility power usage divide by the IT equipment power load.

The PUE performance metric is a step in the right direction – However – It also leaves a lot to be desired. There are many whitepapers on the various aspects, but the biggest question I want to focus on is: What in the Data Center falls into the “IT equipment power load” component of the equation? The Green Grid has left this open to interpretation, and has not released a set of requirements or guidelines for what is included in the PUE.

I personally argue that all equipment in a Data Center should be included (because after all, what we are trying to get is an indicator of the power efficiency of a whole facility), however in practice this is not the case. It is not commonly known or advertised that large Data Centre’s remove equipment that is not classified an "IT Equipment load" purely to improve the performance number. And to make matters worse, when you see a PUE number advertised you have no way of knowing what was included and what was not.

This APC White paper discusses this issue at length, and they point out that the following items in the Data Centre are not typically included in a facility’s PUE:

• Power Distribution Units – According to the APC study, PDU’s lose as much as 10% of the power that they consume, which would have a dramatic impact on the PUE equation if they were included


• In a mixed use facility – The air conditioners and chillers might not be included in the PUE as they are also supporting other aspects of the operation even though the cooling requirements for the actual IT Infrastructure are typically the highest users of the cooling


• An on the side support facility to run and maintain the Data Center – For example a Network Operations Centre (NOC) – For a large scale business with multiple Data Centers, the NOC support infrastructure would be very large and would have a huge impact on the PUE if included
  

As one of my colleagues pointed out “there’s PUE and there’s PUE”.

Don’t get me wrong – PUE is a great first step in the right direction for power performance metrics, but I hope as I’ve shown you there is plenty of scope for improvement.

There are even some industry specific power and performance metrics emerging – For example in the networking world, The Energy Consumption Rating Initiative (ECR) is an emerging performance metric that seems to be gaining traction. The ECR uses a category system, which organizes network and telecom equipment into classes with a different measuring and performance methodology for each class. The classes can then be combined and then the end result is a “performance-per-energy-unit” rating.

APC suggest that a standard set of guidelines should be drawn up which dictates what equipment must be included in the PUE. Personally, I’m not totally convinced that this will help solve the ultimate problem of how to accurately measure and report power performance metrics over all Data Centers with disparate infrastructures and business models.

Which leaves a big challenge moving forwards – How does one build a set of guidelines that can be used to measure power performance over disparate infrastructure solutions and business models?

In my opinion, I believe that the ECR Initiative should be adopted and opened up to all things that use power in the Data Center, with classes of equipment categorized, that all add up to an end figure. Sure it will be a little more complicated, but in the end we would have a performance metric that is consistent from one facility to another and includes and counts all of the power that is used in that facility.

Hope you are all well - I'll be back soon with more Data Center and Windows Server blog posts in the not to distant future!

Regards,

Phil

2009 – The year for consolidation

Hey everybody,
Hope you all had a great Christmas a New Years :- )

I believe 2009 will be the year of cost consolidation for most businesses worldwide. Businesses will look at a wide range of measures to save costs to consolidate and save space (infrastructure footprint), power and cooling costs, infrastructure investment and also staffing costs.

These can be achieved by undertaking a full review of your IT Infrastructure and data centers from the top down.

Some technology which will help with cost consolidation and will be in high demand this year is:

Blade servers – These help reduce the physical space requirements of your infrastructure and compacts the technology into a much smaller form factor. These can reduce the size of your infrastructure footprint by an enormous amount (typically by about half, obviously depending on the size and type of infrastructure you had previously). But most blade enclosures have the capacity to squeeze 8 half height blade servers into 6U’s of rack. If you had 6x 5U servers filling up a rack previously you may be able to squeeze them all into a single blade enclosure, saving half of your rack space.

Because Blade servers are high density, the technology introduces increased power and cooling demands per rack


Virtualization – This is a hot topic at the moment and a hard concept to explain to somebody that hasn’t come across it before. In essence virtualization is a method of having one or more operating systems running in a sandboxed environment on the one physical host. This allows many virtual servers (all servers in their own right with their own operating system, applications and storage) to be run on the one physical host.

Virtual servers are incredibly powerful in a number of scenarios, including:
· Consolidating legacy applications off ageing hardware into a virtual environment – You can take snapshots of a physical server, and turn the snapshot into a virtual server. A classic example is if you need to keep NT4.0 running for a particular application, but have been paying for extended hardware maintenance, you can ‘suck’ the physical server over to a virtual one. This would mean that the server is running on new hardware and would minimize the risk of legacy hardware failure

· You may have servers that have been installed Ad-Hoc for particular projects or applications and at the time little thought was given to running the application on an already established server. Instead a new server was purchased. This over time could lead to a server sprawl where you have racks full of servers which are incredibly lightly loaded, but you have to pay for power, cooling and maintenance for the servers. In this scenario there would be a good case of virtualization where you may be able to get 10 (or more) physical servers lightly loaded onto one large high availability server.


Cloud computing – As businesses look at ways to reduce costs, outsourcing looks fairly attractive. Recent cloud based services such as Microsoft Azure mean that businesses can outsource a whole server function or just a single web application, and with a technology giant behind the service, SLA guarantees are more likely to be achieved with greater reliability and scalability potentials moving forwards.


Some new technology and concepts that will be starting to get some enterprise vision this year include:
- Solid State Disks (SSD’s) – These are high reliability flash drives, that have large capacity options, redundancy options and are quicker and require less power. This is because with the SSD there are no moving platters or drives that need to be powered. To date long term reliability has been questionable, however storage vendors are starting to produce SSD’s that have a mean time between failure (MTBF) of similar levels to that of a traditional hard disk.

- DC Power rather than AC – This is something I will be blogging about a bit later – But the latest data center concept is to supply your entire data center with DC power rather than AC power. With AC power supplied to the rack, the servers have to convert the power to DC anyway, with massive power losses (which take the form of heat) – Thus saving on your cooling at the same time.

- Looking outside the box in terms of your cooling. I have been watching this space now for some time and there are numerous concepts around to help save the power spent on cooling:

o Most data centers are cooled at 68 degrees F (20 degrees C), however a study by Intel suggests that it found no consistent increase in the rate of equipment failure when running the data center at 92 degrees F (33 degrees C). Thus saving power on cooling. There is a catch-22 on this in that the hotter the servers get the harder the fans run, using more power and creating more heat. Dean Nelson (Sun's Senior Director of Global Lab and Data Center Design) suggests in a recent video that there is a "sweet spot" which is probably in the range from 80-85 degrees F (26-29 degrees C).

o Another option that seems to be taking off is if the data center is located in a consistently cool location, consider pumping outside air in to cool the space, and then move the warmed air through to the office space where people are working

Hope all is well.
Cheers,
Phil

Data Centres in a box – What’s the go?

Portable all-in-one Data Centres are starting to become more prevalent. First was the APC Data Centre in a truck, and then came the Sun Portable Data Centre in a container and more recently (and more extreme) a floating Data Centre on a cargo ship.

I’m sure there are more businesses out there who are selling their own spin on the portable Data Centre, but what does it actually mean to have one of these services available to you?

I think that the most common use for these will be as a ‘service’ meaning you rent the ‘box’ for a period of time and contained within that box is a plethora of server equipment, networking kit and some form of portable high-speed internet connection (Satellite?) of which will either come pre-configured or self-configured.

I think the uses of these portable Data Centres are fairly limited however, and are probably confined to a few key areas:
- Disaster Recovery – As long as you can get your hands on one of these portable data centres fairly quickly (depending on business SLA’s), these could be very useful in providing a big punch fairly quickly without the need for costly ‘on standby’ equipment – You just need the space to put the truck, container or ship.
- Companies that have a large requirement to be very mobile – For example a news network, or a TV network covering motorsport, an events company that has a large IT dependency or even a geographical research / materials company (land or water)

I suspect the use for a floating Data Centre would be fairly limited – And probably for quite illegal activities (international waters). Which brings about an interesting question – Who’s laws would the data centre be liable to follow? And who would enforce them?

Anyway – That’s it from me – I’m back from holidays now, so you will start to see some more activity from me over the coming weeks.

Have a good one : - )
Phil