Data centres have become mission-critical in today's information-driven business environment. Unfortunately, they've also become overcrowded, overheated, and expensive. More often than not, expansion is simply not an option.
It's not just escalating powering and cooling costs that are causing concern. The financial penalty associated with lost or stalled business opportunities is also troubling as organisations struggle to find a place to put new systems and technologies that promise to help them meet important strategic objectives.
Faced with little if any available space, companies looking to invest in and implement innovative technology solutions aimed at delivering savings or generating revenue must put such projects on hold. However, by following a few practical guidelines, organisations can extend the life of their current data centres while freeing up space for strategic initiatives and reducing power and cooling requirements.
What are the top problems in the data centre today, and how are these problems becoming more significant?
Space, power, and cooling top the list of concerns for data centres. Just as skyrocketing energy and fuel prices are impacting households across the world, they are also causing a significant rise in the raw cost of operations for the data centre. And market indicators do not yet signal that relief is in sight.
What's more, with new technology continuously being introduced, the problem of data centre space limitation too often stalls important business projects and initiatives. Until organisations find a way to make better use of existing resources, there is simply no place to put new systems.
What should the first step be when undergoing an initiative to optimise the data centre?
At the initial stage of a data centre optimisation initiative, the first step is to understand what is meant by optimisation. We need to clarify what we're being asked to do.
For example, is the organisation looking for a cost reduction or a space reduction? If it's space, exactly how much space is needed, and where is it needed? Does that space need to be contiguous?
Is this a temporary or long term requirement? Is the organisation looking at data centre optimisation as a way to address the company's “green” agenda? Is it looking at the data centre as a vehicle for helping drive profits through new service offerings that will, in turn, support a new segment of business? Is it looking at ways to improve problem resolution, minimise the time to provision a new application, or other similar services?
We also need to understand what tangible results are expected or anticipated as a result of any data centre optimisation initiative.
Understanding the project's key drivers will help data centre professionals set priorities, get results, and earn and maintain support from the business.
What are some of the benefits of data centre space optimisation?
Data centre optimisation can enable organisations to avoid or defer having to relocate their data centres, as well as to revive stalled strategic initiatives, increase the utilisation of existing assets, and reduce power and cooling requirements.
Compared with relocating or expanding the data centre, space optimisation is cost-effective and can produce a fast return on investment (ROI) without a lengthy, expensive consulting phase. Space optimisation can also become effective very quickly, is low risk, and is unlikely to threaten the timing for other projects. What's more, data centre optimisation is an effective way to make better use of a costly resource, and optimisation activities can help facilitate clarity of cost and charge-back by associating each asset with its business owner. A best-case outcome would be to help our clients defer the capital expense associated with a new facility, because we've been able to use an existing resource more effectively.
How can data centres quickly begin to address space challenges?
A variety of activities can help organisations quickly address space limitations in the data centre. For example, organisations can begin to reclaim space by systematically adjusting rack layouts for optimal space utilisation. Where possible, replacing equipment with newer, more efficient equipment will free up space, as will decommissioning and removing equipment that is not being used or no longer meets a valid organisational need.
Organisations can also regain space by consolidating the workload from multiple underutilised servers to a smaller number of servers. Often, there is an opportunity to virtualise server resources, by replacing one or more physical devices with virtual devices that run on shared hardware.
The main point here is that organisations have many choices and they're not all equal. With just a little investment in up-front planning, we can help the IT team evaluate the options and select the approach that is the best fit for their circumstances.
How can data centres quickly begin to address power or cooling challenges?
If power issues are driving the data centre optimisation initiative, establishing “power budgets” and capacity plans will help guide the organisation in reducing its power requirements.
If solving the problem of excess heat is the primary objective, taking a “zoning” approach to managing heat can be very effective. Then, by clearing out unused servers or restacking servers within existing power and cooling tolerances, organisations can quickly begin to see results.
Just having an approach isn't enough. Ultimately, we need to identify and track those key metrics appropriate to the area of interest. We need to acquire and collect the data to determine whether or not our efforts are producing the desired results. Indeed, we need to do it up front to determine whether or not we believe the approach to be a reasonable course of action.
What is the process for determining the top inefficiencies in the data centre?
Once an organisation has identified its objectives and priorities for data centre optimisation and assessed its data centre space utilisation, it should consult with each equipment owner in the data centre-typically the platform leaders, network managers, and facilities manager-as well as with other key stakeholders to understand business needs as well as the role of the company's portfolio of applications, services, and hardware in meeting those needs. The organisation can then leverage this information to determine what equipment must be kept, what should be reassigned, and what can be eliminated.
Needless to say, this is typically not something that can be done in IT's spare time. Indeed, successful data centre optimisation requires the focused efforts of a dedicated team with experience in space-creation initiatives from both a business and an IT perspective. Still, some participation will be required from IT, as they have the detailed knowledge of the environment and the applications.
For this reason, a growing number of organisations are choosing to out-task data centre optimisation to specialists who not only have the requisite expertise but also the resources and time to maintain the sustained effort required to ensure expected ROI. On this basis, the internal IT team members are engaged in a manner that leverages their knowledge and time effectively.
When should an organisation consider completely moving a data centre to drive efficiencies?
Complex relocation projects take time and resources, introduce risk, and involve significant capital expenditures. Consequently, data centre relocation should only be considered after an organisation has thoroughly examined its options for creating more space in its existing facilities-through rack optimisation, equipment refresh, decommissioning, consolidation, virtualisation, or similar efforts-and determines that the supply will not be sufficient to meet the anticipated demand.
Share