Green data centres and high performance computing

ONE TYPE OF APPLICATION that definitely could be moved to a further-afield data centre is high-performance computing (HPC). HPC applications were once limited to scientific research, but many businesses now use HPC to analyse large amounts of data, as well as to create simulations and models. HPC applications are compute-intensive and, when applied at scale, consume large amounts of energy. However, because these applications do not require real-time responses, organisations have complete flexibility over where to locate them. This means that they can now take advantage of lower cost green-energy data centres in new geographies.

Regulatory pressures affecting data centre locations
The data centre industry’s sustainability credentials are well and truly under the microscope, particularly from regulators who are stringently mandating the reduction of greenhouse gases (GHGs) and carbon dioxide emissions. For example, the EU is applying particular pressure via its unilateral commitment to, by 2020; reduce overall GHGs from its 28 member states by 20 percent compared to 1990 levels. This 20 percent reduction commitment is included in the climate and energy package of binding legislation, and it is also one of the headline targets of the Europe 2020 strategy for smart, sustainable and inclusive growth. The EU is so serious about its commitment to reduce emissions that it has even offered to increase this reduction to 30 percent if other major economies agree to undertake their fair share of a global emissions reduction effort.

These regulations use a cap-and-trade model, in which there is a limit placed on the total amount of carbon emissions allowed, with each data centre assigned a carbon allocation. Exceeding this allocation would result in fines and other penalties. The concept of carbon credits and the ability to trade carbon credits is also a part of these laws. There are similar regulatory pressures in the United States. At the federal level, the US Environmental Protection Agency (EPA) requires the mandatory reporting of GHGs. Some states have regulations of their own, such as California’s AB 32, which requires the State to lower greenhouse gas emissions to 1990 levels by 2020. This is the equivalent of taking approximately 15 million cars off the roads. Other examples of US carbon legislation include the Regional Greenhouse Gas Initiative (RGGI) and the Western Climate Initiative. Faced with these increasingly stringent regulations, many businesses are beginning to question their current data centre operations, looking for more environmentally sound alternatives that will keep them out of the sights of the regulators.

Searching for price stability and energy reliability: the cornerstones of a data centre strategy
The volatility of fossil fuel prices, as well as the overall trend of energy becoming more and more expensive is obviously a prime concern for data centres. The incentive to look at sources of energy that offer more price stability is therefore strong and the answer invariably lies with sustainable energy sources. Indeed, in a March 2013 article, Forbes quoted Rick Needham, Google’s director of energy and sustainability, who explained that “while fossil-based prices are on a cost curve that goes up, renewable prices are on a march downward.”

On top of these cost issues, power is not always reliable. Worryingly, the ‘Electricity Capacity Assessment’ report published by Ofgem, the UK energy regulator, in June 2013, warns that, with ever increasing demands being placed on the power grid, the country will inevitably experience tightening electricity margins and face higher risk of blackouts. The prospect of power grid fluctuations – let alone outages – can risk causing major damage to a number of businesses operating in the country. Most notably, these power failures could result in major interruption to the services delivered by essential telecoms, networking and data centre providers, something that has already happened in the US.

Harnessing renewable energy removes these risks. By their very nature, renewables are both predictable and dependable, and offer data centres a long term energy supply that is in stark contrast to those offered by the volatile fossil fuel-powered grids.

Matching the application to the data centre
Once an organisation has decided to adopt green computing resources from a cloud provider or from a colocation service, it will still need to consider which applications are best suited for these environments. In order to make the right choices, businesses will need to have accurate profiles of their applications. One of the most important issues here is latency. For example, interactive applications and stock-trading programs require close to zero latency because users expect quick responses, so it is preferable that they are located in data centres that are physically close to their markets. Organisations running applications that are not latency sensitive will find they are afforded much more flexibility when it comes to choosing a location.

HPC – the ideal application
HPC applications are quite the opposite of flash trading applications. Crunching data 24/7 and requiring very little real-time interactivity, the latency of network connections is not a problem and means firms using these applications have considerable freedom over where in the world they locate them. Today, more and more businesses require the sophisticated analytics, modelling and simulations that are typically performed by HPC applications. But increasing deployments can make life difficult for data centre managers because of the sheer amount of energy these compute-intensive applications consume.

As applications grow, the servers they run on need a commensurate amount of cooling, which adds further cost. This enhances the business case for moving these applications to facilities that are powered by more cost effective and predictable renewable energy sources.

Case study: RMS
An example of a firm that has moved its HPC applications to take advantage of renewable energy sources is RMS, a risk-management solutions company that helps the insurance and financial industries manage risk by providing sophisticated catastrophe-modelling services. Initially sold as a software package, the company found that, as it planned to deliver a new generation of modelling and analytics capabilities to the market, the overall infrastructure requirements became too overwhelming for all but its largest clients. It therefore decided to provide an extensive suite of modern features and capabilities by using a cloud-based delivery model called RMS(one), which is currently scheduled to go live in mid-April 2014.

Initially, RMS will use three data centres that are identical to one another in all but scale. Two are production data centres, located in Europe and North America. The third is currently used to showcase beta versions of the product. Once RMS(one) goes live, this site will be transitioned to being a dedicated disaster-recovery data centre. It will also be used to perform load and stress tests, and to act as the staging environment for deploying to the production data centres. RMS has seen business benefit in sitting this third data centre in Iceland.

Iceland was attractive to RMS for a number of reasons. One was the growing sensitivity of customers, particularly in Europe, to sustainability. Another was the EU directive to reduce carbon emissions. A third reason was, of course, the cost of energy. Iceland addressed all because the power costs are low and predictable and because the power sources are green. These issues were particularly important because the footprint of this site is the largest, due to it being used for disaster recovery, load and stress testing, and staging.

Conclusion
Metropolitan data centres face a battle to prove they are taking sustainability issues seriously as it is clear that the criticism they receive, not to mention the regulatory pressure they are under, will not dwindle any time soon. Nor is it likely that we will ever see fossil fuel prices stabilising or becoming any easier to predict. If firms want to overcome these challenges, they need to adopt new approaches. And, as is apparent from the RMS example, moving latency-tolerant, power-hungry applications such as HPC to locations where they can utilise renewable power supplies makes both financial and environmental sense.