As the rise of big data and real-time predictions heightens CIO interest in high-performance computing, we tell you what to consider–and look out for–before investing in HPC.

There’s growing interest within the CIO community to discover how high-performance computing (HPC) can be employed to create new–and truer–levels of competitive differentiation.

This trend represents a significant shift in the HPC space. For many years, IT leaders placed HPC systems placed in the same category as supercomputers, and as a result, HPC was relegated to large government institutions and scientific research labs.

That’s beginning to change.

More enterprise IT leaders, both in the APAC region and elsewhere, have begun to explore more commercial applications of high-performance computing.

HPC systems are becoming an increasingly necessary ingredient in any industry’s ability to develop new and innovative products.

These forward-looking IT leaders are discovering that high-performance computing is essential in defining new levels of customer experience, risk management, product and service innovation, and unearthing patterns in real-time, among other uses cases.

“HPC systems are becoming an increasingly necessary ingredient in any industry’s ability to develop new and innovative products,” says Steve Conway, Research VP, IDC’s High-Performance Computing Group, in CIO Review.

As companies start to invest in high-performance computing, they are faced with a number questions. Is HPC going to cost me a fortune? What should I watch out for when purchasing an HPC system? Who is the best vendor for my company’s requirements?

To help these companies, here’s a list of guiding principles when investing in high-performance computing.

Principle 1: Control Cost by Starting Small and Scaling Up

Cost is a key factor in any enterprise investment. Unfortunately, high-performance computing suffers from the reputation of being inordinately expensive. While that notion is built around a grain of truth, it isn’t completely accurate.

In reality, high-performance computing need not be investment heavy, especially if you know how much you need. The key is to accurately size the amount of computing power you need you need to start with, buy only those required units, and scale up from there.

The price tag for a small system can start at a $10,000 and even lower.

4 Ideas to Consider When Buying HPC

  • HPC isn’t expensive.
  • You can control costs by starting small and scaling up CIOs should look for value and low TCO
  • Make sure you partner with a vendor that really believes in customer collaboration
  • Domain expertise and innovation should be key investment criteria parameters

However, this strategy requires working with a technology provider that can offer HPC configurations at different price points and performance levels–and that has a modular approach.

One such technology company is Lenovo, whose range of HPC products ensures both price and performance flexibility.

“I believe the Lenovo innovations in the 6U, 12 bay NeXtScale System M5 are a particularly good example, offering configuration flexibility in a very dense package…This design helps customers reduce and manage the complexity…and allows the system to easily evolve to adopt new technologies over time while protecting customers’ investments along the way,” says Karl Freund, Senior Analyst at Moor Insights & Strategy.

At one end of its broad portfolio is the Lenovo Intelligent Cluster, which delivers a factory-integrated, tested and fully supported cluster. Built with Lenovo servers, storage, software, networking, and third-party devices, this solution can help IT departments significantly reduce deployment time, risk, and costs.

At the same time, Lenovo’s Intel® Cluster Ready testing ensures that its HPC systems will run a wide-variety of ISV applications and industry-standard components easily and seamlessly, ensuring IT departments are not locked in.

One the ways in which Lenovo stands out is its water-cooled technology. That’s a key thing, especially from a TCO perspective.
Jonathan Wu, CTO, Data Center Group, Lenovo Asia Pacific

For the most demanding workloads, Lenovo has NeXtScale System, which delivers the density, flexibility, and scale. It also offerings across the price/performance spectrum.

Lenovo’s integrated architecture approach also allows IT departments to easily mix and match air- or water-cooled servers, accelerators, and storage in a dense chassis and use industry-standard networking and racks for significant cost savings.

Principle 2: Look for Value and Low TCO

For most CIOs, price is only one side of the investment coin. The other side is value. What value can I get from an HPC investment? What’s the return on investment I can prove? These are key questions for CIOs.

I believe Lenovo innovations…are a particularly good example, offering configuration flexibility in a very dense package… This design helps customers reduce and manage the complexity… and allows the system to easily evolve to adopt new technologies over time while protecting customers’ investments along the way. Karl Freund, Senior Analyst at Moor Insights & Strategy
Karl Freund, Senior Analyst at Moor Insights & Strategy

Please log in to read the rest of the article. If you do not have an account, click here to sign up.

Jonathan Wu
CTO, Data Center Group, Lenovo Asia Pacific

Based in Beijing, Jonathan leads the High-Performance Computing for Lenovo in the Asia Pacific region. He has over 20 years of experience within the IT industry.