United Devices Founder on Grids, Clusters

By By Derrick Harris, Editor

November 21, 2005

GRIDtoday editor Derrick Harris spoke with Ed Hubbard, founder and chief marketing officer of United Devices, about the company's strategy around SC'05. its recent acquisition of France-based GridXpert (a leader in the manufacturing sector), the issues of Grid standards and Grid adoption, and the differences between Grids and clusters — a topic that arises quite often in the HPC space.

GRIDtoday: How's everything going with United Devices? Will you be announcing any news at SC'05?

ED HUBBARD: Business continues to grow with several of our large customers moving their Grids to a central services model and offering them enterprise-wide.

At SC'05, we'll have a number of announcements, three as a matter of fact. We'll have an announcement with HP, Microsoft and an ISV partner.

Gt: What about the news of the past several months? Can you comment on the GridXpert acquisition? How has it affected the company in terms of customers in new markets and geographies?

HUBBARD: Acquiring GridXpert and its best-of-breed meta-scheduling technology has accelerated a number of our deployments by leveling the complexities of multiple, incompatible schedulers/DRMs. Not surprisingly, we've also opened up a number of conversations within our expanded customer base around the products and services offered by UD. Finally, part of our reasoning for this acquisition was to significantly enhance our position in manufacturing and I believe this has been more successful than we projected. The former GridXpert team (now UD, of course) had a fairly large number of evaluations going on that we've been able to add a lot of value to from a services and technology perspective. I anticipate a very good conversion rate on these evaluations.

Gt: Moving on to the Grid industry as a whole, I'm wondering where you see the market heading in terms of widespread adoption? Would it benefit the cause if the community could agree on some quality standards?

HUBBARD: We continue to architect around open Web services standards, which have served us well with customers. In fact, this is one of the areas where we continue to receive a lot of positive feedback — namely the ability of our customers to easily create application services and portals around our infrastructure in environments like Java and C#.

On the topic of standards, I don't think they've held the industry back as they've largely been irrelevant to date (at least in our deployments). Our view is that this is changing but not in the way you might expect. We actually see more divergence, which was another problem we wanted to address with the GridXpert acquisition. By embracing the reality of the market (multiple, incompatible schedulers / DRMs) and providing a solution to eliminate that unnecessary complexity for our customers, we believe we have a superior product set to solve what our customers really face in their everyday production environments.

I don't think this is going to change very quickly since there is a huge gap between public/press perception and commercial reality that needs to be closed with some open conversation:

  • There is a perception out there that WSRF/OGSA & Globus are standards.
  • The reality is that Globus is merely a Grid software toolkit. In other words, it is an implementation geared toward academic environments where it has been somewhat successful — it is not a standard by any means. OGSA is really WSRF and is currently under review at OASIS and it does not address current enteprise issues around interopability. Most of the requirements addressed in the current document reflect academic needs of running multiple Grid services across different organizations.

We've tried hard to stay out of this debate and just be customer-driven, but this is the core problem of standards for the Grid market that no one seems willing to admit or discuss openly.

Gt: How is it that United Devices is so successful in the Grid space given the somewhat disappointing enterprise adoption rates? Does it have something to do with simply targeting the right customers in the right industries?

HUBBARD: We have 100-plus customers today ranging in size from 20 nodes to 3 million, but this hasn't been an overnight success. We have worked hard and slugged it out with more than a dozen other players that have come and, thankfully, gone. This has left a pretty stable set of solutions that have settled into that age-old competitive triad (UD, DataSynapse and Platform) where there are three quality solutions in the market that all own their respective verticals and compete in the verticals where no player has a clearly dominant position.

I would argue that Grids have already penetrated the majority of the Fortune 500 in some form even if they haven't been adopted as an enterprise-wide service yet. For UD, we chose to focus early-on on life sciences and have been very successful in that vertical market. We've built up a ton of credibility, a high-quality set of references and large, complex deployments that I'd put up against any competitor's reference set. With GridXpert, I hope we've successfully telegraphed that we plan to repeat this same success in the manufacturing and other environments and, in fact, are already well on our way.

Gt: On to Supercomputing … what is it about this show that draws so many attendees and exhibitors? For how long has United Devices been exhibiting at the show?

HUBBARD: I think the show draws well because of the natural progress of technology, e.g. what used to be a highly specialized supercomputing application 10 years ago is now running on a Grid and the results can be visualized and manipulated on a single workstation. The democratization of these engineering analyses has continued unabated and the show where you can go to discuss them across the board remains the SC'XY shows.

UD has either exhibited or attended the show (based on our strategy for the given year) since 2000.

Gt: Seeing as how United Devices operates in both the Grid and cluster markets, which do you tend to focus on at SC events, which seem to focus a more on the cluster/HPC market?

HUBBARD: We typically focus on clusters at the SC'XY shows for both practical/logistics reasons and because we just think its a better fit from an application perspective. If you look across the set of applications in use and demo'd at the SC'XY shows, you'll find a large number of MPI-based applications that really require the speed and low interconnect latency of a cluster to run effectively. A Grid is simply a set of resources — these resources could include clusters. We don't see a Grid market as different from a cluster market. We can effectively manage clusters stand-alone or as part of larger Grid deployments through our Grid MP platform (MPI jobs included).

Gt: Finally, and I'm starting to ask this question a lot because I feel that there is still some confusion out there, can you define, at least from a United Devices perspective, the differences/similarities between clusters and Grids?

HUBBARD: Grids are the superset in our vernacular. A Grid can be made up of any available device that may be useful to the application type(s) the enterprise is trying to accelerate, raise the reliability on, etc. That said, we love starting somewhere on a cluster because we can clearly articulate how this single, little cluster could grow up to be the enterprise's Grid — we call this a “Grid-ready cluster.” Many of our current implementations started out as a cluster and built a Grid of clusters and other resources.

Clusters are, in general, simpler animals. They are typically self contained, built from homogeneous hardware, have dedicated, private-switched networks and storage, are managed carefully by dedicated staff and are often a better path from a price/performance perspective for a number of applications.

Grids, on the other hand, often include a cluster, or multiple clusters. Sometimes running the same underlying scheduler/DRM, but sometimes not. They can also contain servers that are part of “'hot” recovery sites, underutilized machines from, for example, a company's Web farm that do very little at night but which can add significant power to a Grid with high overnight processing demands. Workstations and PCs are also often a part of enterprise Grids and this adds complexity around security requirements, data management and caching, etc., that we've tuned for in our Grid MP platform. Finally, Grids become a much larger, integrated part of the business's core value creation and as such absolutely require a higher level of manageability than a single cluster. This area is one where we have been increasingly focused over the last couple years driven by our customers.

The Grid is fundamentally about removing the current hard, static binding between an application and its execution environment. In the new world, this binding will be soft and very dynamic. The benefits of doing this are very significant for our customers.

Gt: Is there anything else you'd like to add about United Devices, SC'05 or the Grid market in general?

HUBBARD:
From both introspection and looking at our competitors, we fully believe that Grids are moving into mainstream adoption. All indications from our current customers and prospects point in this direction.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire