Transforming IT Blog
Join us in the Transforming IT HP Blog where we will discuss reinventing IT to overcome obstacles and take advantage of Instant on Enterprise opportunities.

How Big Data Leads to Bigger Energy Savings for the Data Center

The term “big data” is being used quite a bit lately. Without proper context, it becomes yet another throw-away term like so many others that the technology industry just loves to invent. However, when it is put into an energy context, it takes on a very legitimate and important connotation: the ability to take hold of and analyze large amounts of information and give back meaningful, concise data that can be used to make decisions regarding energy efficiency measures.


This is an important corollary that must be grasped: in and of itself, big data is meaningless. It is the ability to exploit big data that leads to successful outcomes.

 

Here is an example to help illustrate this point. Let’s follow the chain that drives energy and water efficiency in data center facilities. This chain runs from the innards of the computers in the data centers, all the way to the source of electricity generation. This chain has many links, and even some offshoot chains that lead to auxiliary functions. Along this chain, at every link, data is developed. These data need to go up and down the chain, analyzed and synthesized to form higher level data sets. But all of this is being done in an iterative fashion, using milliseconds as a time scale. It becomes apparent that this process, running 24/7, generates massive amounts of data, which is required for instant, real-time decision-making, but also for mid- and long-term use for making decisions in the future.


So what are these links and what are these types of data that are generated and used in such a fashion that they are entitled to a special moniker? While I can’t claim to give the answer to this question in its entirety, as a mechanical engineer having an area of expertise in data center energy efficiency, I should be able to cover most of it.


As the computer workload throttles up and down during the day, the electricity that is consumed also varies. Also the type of workload that is being run may have a lower reliability requirement allowing for higher inlet air temperatures. This will then require the HVAC system to adjust accordingly, which also changes the electrical consumption. This is an example of a loop that will need many iterations in order to optimize energy use.


Concurrent to this, as the time of day changes, the electricity rates may change, so it might be advantageous to move the workload to a different data center.


All of these scenarios generate tremendous amounts of data that need to get aggregated and analyzed. But these are just a few simple examples.


Now consider the electrical utility’s perspective. The utility is looking at the types of data mentioned above but from a much higher level, since it will be aggregating and analyzing data sets that are many orders of magnitude greater than a single facility. Using this data, the utility will be able to understand in a much more dynamic, just-in-time fashion when and where electricity is needed. This allows for a more accurate prediction of the base load and when additional electrical generation sources might be needed. (This is essentially the same model that is followed now; the big data will enable much more timely information and rich historical data).


So the bottom line is that all the way along the chain, from processor to power plant, data is used to optimize not only the single component, but all other components in conjunction with it, in order to develop the highest level of optimization. If properly managed, this is how big data can lead to bigger energy savings.

 

Learn how HP Data Center Transformation Services can help you grab the best data center opportunities.

Read how HP storage solutions helped one company cope with increased data volume without adding IT staff (369 KB, PDF).

Comments
Renewable Energy News(anon) | ‎06-01-2012 04:41 PM

This is very informative. Thank you for sharing.

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the community guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
Showing results for 
Search instead for 
Do you mean 
About the Author
Bill is the Principal Data Center Energy Technologist for HP Technology Services. Kosik is a licensed professional engineer, LEED Accredited...


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation