Can ‘always on’ reliability co-exist with Green IT and other energy saving initiatives? Not surprisingly, analytics and business intelligence deliver the answers. Companies like American Power Conversion have online calculators to help decipher what savings are possible and how to measure performance while maintaining the desired level of computing power, cooling costs and other overhead such as security or load balancing.
Faced with an either/or problem, a typical answer is “It Depends.” The units of measure, periods of time or other factors you choose to define the problem will all affect the answers you get. I’ve seen the confusion in my own “data center” – a home office with three PCs. My own analysis of utility bills show the monthly amount paid is not yet declining significantly although my usage has cut the total number of watts consumed.
For more on what kinds of data and measurements are most useful in balancing cost saving versus performance levels, Mission Critical magazine tells us that often today’s data center energy calculations are based on old (2005-era) assumptions — creating problems for data center managers planning future installations. A look at your own experience and costs, usage, efficiency is worth a lot more than general, outdated data or generalizations that may not apply to your situation.
Analytics allow you to use as much of your own data as possible and choose a consistent, defensible metric, compared monthly, quarterly or yearly. Then compare the historical trend using ongoing plans for improvement – better is never the enemy of best when it comes to making ongoing small changes. Business intelligence tools can monitor these baby steps and show you how far you’ve gone — showing big moves over time.
Using analytics to track patterns and spot anomalies, or explain why changes occur is vital. Maybe a manufacturing plant had to run overtime and had higher energy costs – were they offset by the profits from additional sales? No data exists in a vaccum. There are always explanations for why something happened and then additional details to show that perhaps that anomaly was worth it (or not) and those are the judgment calls that business intelligence can accelerate and improve.
After all, knowledge is power and power should be used efficiently, not wasted.
David Wallace
Spotfire Blogging Team
Image Credit: Microsoft Office Clip Art
