In the data center, inefficiency can be measured in terms of lost time and wasted money. In a recent survey of 200 data center managers equally split between the US and UK, DCIM adopters confirmed that this class of tools makes their work “easier, faster, and saves money.”
If DCIM can truly save time and reduce operating expenses, why did the survey find that 43 percent still rely on manual alternatives to DCIM tools? What’s holding back DCIM deployments?
No time or money for DCIM
Intel commissioned Redshift Research to find the answers to these and other questions about the current approaches to data center capacity planning and forecasting. They discovered that MS Excel remains a popular tool and almost 10 percent of data center managers still pull out a tape measure for any expansion or layout changes. Just over half – 55 percent – are able to benefit from DCIM tools.
Surprisingly, the manual approaches are not more typical of smaller data centers. Even among the larger data centers with more than 1500 servers, the 43/55 split held true (with two percent unsure about or failing to carry out any planning or forecasting).
What were the two most common reasons for not deploying DCIM? Almost half of the non-DCIM segment (46 percent) cited the expense of the tools as an inhibiting factor. A further 35 percent indicated that they lacked implementation resources.
The Catch
The lack of budget combined with a lack of resources and time to deploy might initially seem like deal breakers for DCIM. However, the survey questions went deeper to discover the rest of the facts around DCIM investments and results:
- A lack of DCIM correlates to wasted time. More than half (56 percent) of manual planners need to devote more than 40 percent of their time every month to capacity planning and forecasting.
- DCIM tools enable optimizations that trim operating expenses. Of the DCIM users, 63 percent reported using DCIM analytics to improve cooling efficiency. These users were also more likely to conduct hotspot audits and CFD simulations.
- Without DCIM, 32 percent lack actionable data. These data centers struggle to deal with common constraints such as limited floor space and power.
- DCIM tools enable quantification of outage costs. 72 percent of DCIM users knew the cost of outrages, compared to only 14 percent of those without DCIM. The survey results included an estimated average cost of $28,900 per outage, which remains a hidden cost for data centers that lack quantification capabilities.
- Twice as many DCIM users reported fast recovery times. Respondents reported that recovery time after an outage averages 7 hours and 53 minutes. Of the DCIM users, 21 percent said they could recover in less than 2 hours; only 11 percent of the other data centers could achieve this recovery speed.
Conclusions
Data centers face many complex challenges today, and must continually prioritize to balance the demands of users with budget, staff, and facilities constraints. The recent survey results make it clear that DCIM is helping data center managers by providing access to valuable information, identifying problems, and helping uncover the true costs, implications, and causes of problems such as outages.
Without DCIM and increased levels of automated control, data center managers must spend increasing amounts of time and inevitably lack adequate information for the most common challenges around capacity planning and allocations.
Research into DCIM solutions and cost justifications should therefore take into account the proven longer-term benefits, or data centers face the risk of short sightedness. Short-term costs, in terms of both time and expenditures, should never be ruled out if they can yield much greater savings going forward.
To read more about the survey and results, download a copy of the Intel white paper: “Inefficiencies cost data centers time and money.” Follow Intel DCM on Twitter.