August 17, 2007
In many instances, supercomputing solutions built for customers are firsts, fastest, largest or most unique.
In June, IBM proclaimed Blue Gene/P as the world's most powerful supercomputer (three times more powerful than its predecessor). In July, the University of Reading announced the UK's largest academic supercomputer and Germany's Jülich Research Centre announced plans to build the world's second-fastest supercomputer. Just this week, the US National Science Board awarded $208 million to the University of Illinois to build a petascale supercomputer, which would make it the new world record holder.
When you're working on the leading edge in this way, problems and hold-ups can and do occur. In this case, it is essential for customers to know that the main players in any supercomputing project -- the integrator and the primary hardware vendor -- have a good relationship that ensures projects are run as efficiently as possible.
It is imperative that customers should look for evidence of this relationship when selecting partners for their supercomputing solution.
1. Long-term commitment
When delivering supercomputing solutions it may take six to nine months before a project is fully complete -- from conception to installation and configuration. The hardware vendor and integrator therefore must have a long-term commitment to each other.
This can only be achieved if an integrator holds a relationship with just one or, possibly two, primary hardware vendors and does not maintain a policy of "playing one vendor off against the other." An integrator and primary vendor should share plans, business goals and build their businesses together. In addition, an integrator should have a policy of wherever possible building supercomputing solutions on the server and storage technology of just the select vendor only.
An integrator working in this way will win business because customers have confidence in the strength of its partnership with the vendor and can see the benefits that this brings to them.
Finally, supercomputing solutions are also often sold to customers alongside commitments of future product direction. This highly sensitive roadmap information will only be made available to an integrator if it can demonstrate commitment to the specific vendor.
2. Interlinking partnerships
For complimentary technology -- supercomputing interconnects, switching or cluster management software for example -- it is often the case the primary vendor will not supply this technology. After all, no one company can supply everything. However, a good integrator will look to build partnerships with complimentary technology vendors that also share partnerships with the integrator's primary technology vendor. These interlinking partnerships will help to build and develop the overall relationship between integrator and primary hardware vendor.
3. Regular interaction
Customers should look for evidence of regular interaction between vendor and integrator. Vendor and integrator should interact at multiple levels within each others' organisations. It is also essential for a vendor to provide an integrator with a dedicated account manager -- someone who has a focus on that integrator and its customers. This human interaction helps to ensure "the cogs mesh" and "don't grind" between each organisation.
4. Innovate together
Integrator and hardware vendor should be committed to jointly innovating and researching new supercomputing, High Performance Visualisation and High Performance Storage solutions.
An integrator will often have a number of people constantly engaged in researching their own new and interesting supercomputing solutions and, in addition will be further investigating the research of the primary hardware vendor.
5. Partner Programmes
A vendor should demonstrate commitment to its integrator through modern partner programmes. Partner programmes enable a vendor and integrator to decide how best to open markets, build specific go-to market plans and clearly share the risks and rewards. Importantly, the programmes helps a vendor and integrator present a shared view or outlook to the customer.
6. Single team
When installing new technology for the first time, customers should also look for evidence that both the integrator and hardware vendor will be available on site throughout installation. This should be a regular feature until knowledge of a new product is completely transferred from vendor to integrator.
Ultimately, vendor and integrator should demonstrate to customers a commonality of interest -- a business need that strongly links them both together.
A large hardware vendor such as IBM will have literally hundreds of thousands of customers and will find it increasingly expensive to services those customers. It may also find that smaller customer sites are not used to, or willing to work with large global organisations. A vendor will rely on an integrator that can demonstrate market knowledge, technology know-how, a strong backing of customer references and the ability to integrate products without damaging the brand of the vendor. A vendor will also benefit if an integrator has its own profile and brand awareness.
In return, an integrator will rely on its vendor to continually develop products and solutions, but importantly offer good technical support and pricing. A vendor can also help an integrator to secure sales and win deals by ensuring the success of its own brand.
-----
About the Author
Julian Fielden is Managing Director of OCF plc. Responsible for the strategic direction of the business and its day-to-day operations, Julian oversees all customer accounts and relationships with key technology partners, IBM, Microsoft, Fakespace, Voltaire, Intel and AMD. He was responsible for establishing OCF plc in January 2002, purchasing the business and assets of OCF Limited in a management buy-out.
Julian is a Chartered Accountant, having trained with Ernst & Young in Leeds, UK. After a brief spell working for Ernst & Young in South Africa, Julian returned to the UK becoming a partner in a Leeds-based accountancy firm. Julian left the accountancy profession in 1990, joining an Industrial group as Group Financial Director. During his 6 years there he experienced a broad range of corporate activity including acquisitions, disposals, company reorganisation and company flotation. Julian then entered the world of IT in 1996 joining a predecessor firm of OCF plc, Open Computers and Finance.
Julian is a Fellow of Institute of Chartered Accounts (FCA) and member of the Institute of Directors (IoD). He has a degree in Economics from Swansea University.
Jun 19, 2013 |
Supercomputer architectures have evolved considerably over the last 20 years, particularly in the number of processors that are linked together. One aspect of HPC architecture that hasn't changed is the MPI programming model.
Read more...
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Read more...
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Read more...
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Read more...
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Read more...
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.