Data Center Infrastructure Management (DCIM) is one of those product categories that sounds too good to be true, but cost and complexity prevent most organizations from seriously considering it as a near-term initiative.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=iIt’s kind of like the prettiest girl in school who doesn’t go to the prom because none of the guys can get up enough courage to ask her.
While virtually all data center operators are eager to drive greater efficiency in their infrastructure and more than half utilize a variety of power management tools, according to a recent Intel/Dell report, relatively few have pulled the trigger on an end-to-end DCIM suite. The reasons run the gamut from complicated deployment and integration issues, potential disruption to data processes and uncertain ROI on the back end. And at the moment, the need for increased scale and flexibility within the data center trumps driving down the energy bill, at least until you start to contemplate hyperscale and hyperconverged infrastructure.
This is leading to a fair amount of turmoil in the DCIM software market. Last fall, IDC reported that companies like Schneider Electric, Emerson Network and Power and Nlyte were sitting on top of the market but were bracing for new competition by a range of start-ups and well-heeled players in the power industry like ABB and Siemens. Since then, however, activity has been mixed, with Nlyte acquiring FieldView Solutions to round out its real-time monitoring and analytics capabilities, and CA, which had provided DCIM solutions for Facebook and other data giants, pulling out of the market entirely because of a lack of sales momentum among mainstream enterprises.
For most organizations, then, the question is not whether to embrace DCIM, but how to do so effectively. To that end, Schneider’s Matthew Baynes says DCIM is no different from any other class of software, so the disconnect between expectations and reality will start to ebb as experience gains. For one thing, DCIM can only manage what can be measured, so loading up on software before investing in infrastructure visibility is a sure way to diminish your returns. As well, too many early adopters went for the forklift upgrade in the hope of transforming aging, disconnected systems into dynamic, energy-sipping infrastructure. But trying to do too much too soon will most likely bog the entire process down in complexity and confusion, and then inevitably lead to disillusionment.
The better way to approach DCIM is gradually. While no one wants to be forever stuck with a mish-mash of monitoring, control and automation systems, the fact is most legacy infrastructure has evolved piecemeal over the ages and therefore consists of a wide range of management solutions. Layering most or all of these systems on a common monitoring platform is a first step toward a cohesive view of the data environment, and once that is done, individual management systems can either be folded in or replaced with a more unified approach.
But this isn’t going to happen overnight, or on the cheap. And with energy costs on the downswing at the moment, the immediate ROI will probably not be as good as it would have been a year ago. Of course, these costs have a tendency to climb just as quickly.
For most organizations, however, full DCIM will likely be more of a consideration in greenfield infrastructure deployments where you have the luxury of building advanced power management from the ground up. But don’t think that this can be done as an afterthought once the workloads have been deployed. Power consumption and the ability to move data in accordance with energy patterns within the data center needs to rank high in the list of criteria for any new deployment, or else you wind up with the same mismatches that are driving up costs on current infrastructure, except this time it is on hyper-dense, hyperscale hardware.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.