Faster Time to Insight

Earlier this summer, prospects looked good for the UK manufacturing sector. In June the second quarter 2011 “Manufacturing Outlook” survey was published by EEF, the manufacturers’ organisation and BDO LLP the world’s fifth largest accountancy network. The survey showed that UK manufacturing had expanded in six consecutive quarters. It also found that positive investment and recruitment intentions were signalling a gradual increase in confidence. It forecast solid growth through 2011 and 2012.

Just three months later, the situation looked far less rosy with manufacturing in August shrinking at its fastest pace in more than two years as export orders plummet. Senior economist at Markit, Rob Dobson, commented, “the second half of 2011 has so far seen the UK manufacturing sector, once the pivotal cog in the economic recovery, switch into reverse gear.”

If the manufacturing and engineering sectors are to overcome this worrying development and continue to lead the UK into economic recovery, high-quality design will have an important role to play. The key issue for large engineering companies that design in the UK is how to leverage the latest technologies to support their design work and high performance computing (HPC), the use of parallel processing for running advanced application programs efficiently, reliably and quickly, is an excellent example of that.

The primary benefit of HPC is its ability to drive ‘time to insight’ – the length of time taken between the presentation of the problem and an understanding reached of how to solve it. By shortening this cycle, manufacturers have the opportunity to compress the design cycle, add time to the build space and potentially reduce time to market.

Challenge Assessment

Engineering companies face a range of challenges in their bid to achieve effective HPC. Cost is inevitably a key factor.

Organisations typically want bigger and better computing systems, particularly if they are running key HPC applications like computational fluid dynamics (CFDs) or finite element analysis (FEA). Yet, the size of computing resource and associated cost of running such applications can be prohibitive.

The problem can be further exacerbated by the tendency for different groups of HPC users within the same company to run their own siloed HPC systems. The core dilemma for manufacturers is how can they achieve an implementation comprehensive enough to cover all of their requirements but sufficiently consolidated to be run in an efficient manner.

Finding a Simpler Solution

To be successful in the future, HPC needs to move away from a focus on ‘exotic’ proprietary architectures towards a more open standards–based approach. Currently, HPC is still trying to get up to speed with the kind of standardisation seen elsewhere in the ICT space.

To accelerate adoption of the technology, vendors first need to make systems easy to learn and use. By so doing, they will help drive their customers to adopt scalable and flexible architectures. They can help streamline this process not only by implementing open, standards-based solutions but also by using straightforward commercial off-the-shelf technology (COTS).

Second, providers need to be prepared to work with their customers in the manufacturing and engineering sectors to ensure that their software and applications are modified to the point where they are capable of leveraging the size and scale of computing systems that HPC typically has to support.

Delivering On Demand

Running in parallel with these developments we are also seeing the emergence of a completely new delivery model for HPC, which Bull has helped to pioneer, known as HPC on demand. This approach involves the solutions provider investing in infrastructure that gives prospective users the opportunity to buy access to that computing resource rather than having to make an upfront investment in a complex IT hardware implementation.

HPC on demand addresses the fundamental difficulty that many engineering companies have in managing their requirement for IT systems, namely their inability to accurately anticipate their future workload. In most cases, there will be no clear or predictable pattern of need. If a given company implements HPC in a traditional way, they would run the inevitable risk of either having to vastly overprovision or be continually limited in what design work they could do because they did not have sufficient equipment in place. The simple way to overcome that is to meet peak requirements on an on demand level.

Manufacturers and engineering companies adopting this new approach no longer have to worry about the complexities of running their own environment or the negative impact that doing so might have on their profit and loss account. Instead they can tap into the available computing capability across the web as and when required.

The emergence of HPC on demand is also opening out the technology to a broader range of companies including many SMEs in the manufacturing or engineering spaces who have basic computing resources, but cannot invest further and would rather seize the opportunity to outsource hardware and technical support so they can concentrate their own resources on their core business.

Freed from the shackles of technology, such organisations now have complete freedom to innovate: with HPC on demand, they have no need for physical infrastructures and no more maintenance requirements, while still having access to high-performance tools for innovation.

Looking Ahead

The right delivery model, whether it is one based on an open standards-based approach or one founded on full HPC on demand, helps manage this power to deliver advantage for manufacturer, businesses and the overall national economy. With the latest UK manufacturing data far from positive and the tentative economic recovery beginning to falter, this capability can be key in ensuring manufacturing and engineering companies have at their disposal every possible means of driving operational efficiency and competitive edge.