Simple processes that can help make the data centre more energy efficient, according to Dean Riach, of GlassHouse Technology (UK). How senior managers can encourage data centre managers to introduce energy efficient policies.
Dean Riach, Contributor
Consolidation and centralisation activities are prominent within most large businesses today; not only because of the major cost savings and efficiencies they can deliver, but also because they are deemed as data centre (DC) green initiatives.
There’s no doubt that the data centre carbon footprint is continuing to grow massively; driven by the proliferation of technology to inform information hungry end-users and the continual flow of feature rich applications. However, managing energy efficiencies has been seen to have limited success to date - and this isn’t surprising because of tight operating parameters and underpinning policies traditionally focused on end user quality of service and availability.
There are a number of ways that a business can ensure their consolidation projects work harder for them and reduce carbon emissions:
- Defining a set of processes which would begin to make a data centre more energy efficient need not be complicated or time consuming in achieving early measurable benefits. By understanding the drivers and then identifying the appropriate Key Point Indicators (KPI) on which efficiencies will be measured is an essential first step and a combination which is often overlooked.
By working with the business and applying KPIs to planned and ‘in-flight’ initiatives such as virtualisation, centralisation or cloud, cost savings can be readily identified. Doing this provides much needed visibility to the data centre manager on which to develop a supporting DC strategy.
- Interface with the business early on in the process. Any architectural decisions should be aligned to physical energy savings and the operating model required in supporting business requirements. Tangible savings within the infrastructure may not equate to overall savings where the solution is not optimal for the business or support model and may increase the wider carbon footprint.
- Any reference target architecture must incorporate a set of known measurable attributes from which the data centre manager can base rack layouts, cooling and power. By identifying and consolidating key attributes which map both business and technical elements into a central view, this will expedite the route to aligning low energy devices to a long term strategy.
- Implementing virtualisation strategies should not focus purely on the physical elements. Interfacing early with the business application owners and users will provide much needed time to encourage the use of shared infrastructure which is often a stumbling block in achieving overall cost savings.
- The sourcing of low energy devices should include environmental directives regarding construction and ultimately recycling, the latter of which may require specialist considerations.
In summary, the key objective is to develop the basis of a simple framework, which can both achieve measurable efficiencies very early on, and identify where these savings are being applied. The ability to develop an efficient architecture based on the alignment of business and technical requirements will not only accelerate the realisation of energy efficiencies - and cost savings - but will provide a much needed link between the business and data centre now and in the future.
Dean Riach is the Data Centre Practice Lead for GlassHouse Technologies (UK), where he brings over 20 years experience within application development, IT Operations and services. Dean joined GlassHouse from Accenture where he was the IT Service Management practice lead responsible for the delivery of multi-million pound complex programmes.
Dean’s previous roles covered business and technical management within HDS, EMC and BT where he used his broad range of application, operational and technical experience in providing leading edge services. Dean spent his early career with EDS leading the development and deployment of high transaction lottery database systems deployed within Scandinavia and the US.
No Comments »
No comments yet.
Leave a comment
You must be logged in to post a comment.