Historically, for a data centre to ‘meet new needs’, it would simply add floor space to accommodate more racks and servers. However, the demands for increased IT resources and productivity have also come hand in hand with increased need for higher efficiencies, better cost savings and lower environmental impact. Third party colocation data centres have increasingly been looked at as the way to support this growth and innovation, rather than CIOs expending capital to build and run their own on-premise capability.
High Performance Computing (HPC), once seen as the reserve of the mega-corporation, is now being looked at as a way to redress the IT budget/performance dichotomy and is requiring data centres to adopt high density innovation strategies in order to maximise productivity and efficiency, increase available power density and the ‘per foot’ computing power of the data centre.
The industry view of high density in the data centre
Source: Thinkstock / Eimantas Buzas
Industry views around high density vary widely. Data centres built as recently as a few years ago were designed to have a uniform energy distribution of around 2 to 4 kilowatts (kW) per IT rack. Some even added ‘high density zones’ capable of scaling up if required, but many of these required additional footprint to be provided around the higher power racks to balance cooling capability, or supplemental cooling equipment that raised the cost of supporting the kW density increase.
There are still many differing perceptions around what a high density zone might mean in terms of capability and cost to the end user. Moore’s law states that computer processing power doubles every year, which theoretically means that the capabilities of data centres year on year could vary quite significantly depending on when they were built.
Gartner recently defined a high density capability as one where the energy needed is more than 15kW per rack for a given set of rows, but this is being revised upwards all the time with some HPC platforms now requiring performance in the 30-40kW range - sometimes referred to as ‘ultra high density’.
Who can support HPC and why is it important?
Being able to support High Performance Computing in the data centre, using high density innovation, has become the next battleground for colocation providers and this goes some way to explaining the differing views around what high density actually is – and how to support it. Some will have higher density capabilities than others – though few providers will admit it.
High density capability will be extremely important for businesses deciding which third-party data centre to use in the future. If high density has been designed ‘in’ from the beginning, it provides the ability to support the next generation of businesses IT infrastructure for HPC, thus optimising the data centre footprint required and the overall associated costs.
This means that irrespective of whether existing data centres take steps to offer high density, they are playing catch-up with a next generation of intelligent data centres that already have this capability.
As a result, data centres over five years old will come under increasing pressure to align to new, more powerful technologies being installed in the data centres, if they want to remain competitive in the marketplace.
Upgrading legacy data centres for high density
Existing data centres that do take steps to offer high density and accommodate the installation and running of HPC will have to upgrade their facilities in most cases. This however, is easier said than done.
Although the concept of high density is straightforward, it involves a lot more than simply main-lining more electricity into the building. Therefore, it’s essential that before a data centre can support this requirement, it has a robust and fit-for-purpose infrastructure in place.
High density not only requires increased quantities of power per cabinet, but also next generation cooling capabilities, which are extremely difficult to retrofit.
Advanced cooling is essential as more energy consumption and harder working servers naturally equate to more heat. Given data centres have very strict operating parameters when it comes to temperature, this means that a data centre needs not only to be able to cope with the extra power being piped into the building but also the extra heat being pumped out.
Consequently, many traditional data centres struggle to provide high densities within the racks – even at a medium density – without a supplementary cooling and support infrastructure or a compromise to the original data centre design, which would be extremely costly and negate the customer cost benefits of HPC.
Make the right high density choice
Businesses need to understand that making the right choice is not simply about the data centre, it is also about making the right HPC platform choice.
Lots of data centres will claim to deliver high density computing, and technically speaking, many will, but only intelligent data centres that have been built from the ground up with high density in mind will be able to do so cost-effectively.
As such, it’s more important than ever that businesses conduct due diligence before signing up with data centre providers otherwise they run the risk of tying themselves into costly long-term contracts that neither meet the current or future needs of the business.
Matthew Larbey is the product strategy director at VIRTUS Data Centres.