Appro CEO Shares HPC Vision

By Tiffany Trader (HPC)

November 6, 2008

In this interview, Appro CEO Daniel Kim describes how Appro has been addressing the needs of high-performance computing customers worldwide to do more with less. He also provides a glimpse into Appro’s vision and opportunities for its supercomputer and high-performance cluster solutions.

Appro has announced a number of high-profile customer wins, particularly with the three U.S. National Labs. Can you recap some of these wins and explain their significance?

Daniel Kim: Appro has been working with Lawrence Livermore National Laboratories (LLNL) for about four years. LLNL is one of Appro’s top customers for supercomputing projects. Appro has installed several cluster solutions at their site, reaching an approximate total peak performance of 480 Teraflops of supercomputing power.  These Appro clusters represent the largest Linux clusters installed at LLNL today.

More recently, we have been working with all three National Nuclear Security Administration (NNSA) National Labs — Lawrence Livermore plus Sandia and Los Alamos — as part of the Tri-Lab Linux Capacity Cluster (TLCC) program. Under this TLCC subcontract, Appro has provided nine clusters of various sizes ranging from 144 nodes to 1,152 nodes with aggregate peak performance of 620 Teraflops and nearly 97 aggregate Terabytes of memory. These clusters are being used in the Advanced Simulation and Computing (ASC) program and the National Nuclear Security Administration’s (NNSA’s) Stockpile Stewardship program. This Appro contract is significant because it marks the first time that the three laboratories have teamed up to purchase and deploy a number of systems sharing a single architecture design.

Also this year, Appro delivered a 95 Teraflop Appro Xtreme-X™ Supercomputer to the Center for Computational Sciences at the University of Tsukuba, one of Japan’s leading academic research institutions. By the way, the Xtreme-X supercomputer is now listed as the second-fastest system in Japan.

Finally, Appro has also completed the delivery of a 38 Teraflop Xtreme-X Supercomputer to upgrade the compute infrastructure of Renault’s Formula One team. The powerful Xtreme-X2 system is installed at the brand-new ING Renault Formula One (F1) Computation Fluid Dynamics (CFD) Centre in the U.K. The supercomputer is being used by the Renault F1 Team to run full-car simulations for their 2009 racing cars.

These multimillion-dollar agreements show that Appro can design, integrate and deliver supercomputing solutions based upon customers’ demanding individual requirements.

Why do you think Appro is finding such success with these and other new customers?

DK: Appro is applying its experience and expertise to design industry-leading supercomputing solutions that enable customers to do more with less, with a reliable and scalable infrastructure that offers a good return on investment.

Appro offers performance, reliability, flexibility and choice, so customers can select high-value supercomputing solutions that meet their technical requirements. We deliver open product architectures that improve bandwidth, performance scalability, reliability and availability while reducing latency and bottlenecks, to ensure the performance of customers’ applications.

Appro’s engineers are focused on making sure that our supercomputing solutions deliver the processing, memory and networking performance needed to accomplish even the most demanding HPC tasks quickly, reliably and affordably. Our optimized infrastructure provides a tremendous amount of added value for customers looking to accelerate their business growth.

What are the benefits of Appro’s Scalable Units design for installations (such as the three U.S. National Labs) with extremely high-capacity computing needs?

DK: Appro’s Scalable Units (SUs) are single hardware design points that enable multiple clusters to be built, based on the same architecture. One of the benefits of Appro’s SU design is that volume purchases of the SUs can achieve economies of scale that rival purchases of a single large system, but with all the flexibility and other advantages of HPC clusters.

Each SU can be used as a highly replicated unit to build clusters of different sizes, depending upon programmatic requirements. Systems ranging in size from 1 SU to 16 SU are possible, although the largest one delivered at LLNL was an 8 SU cluster with 162 Teraflops. Because all the SU-based systems use the same architecture, Appro’s system integration and deployment of these clusters also is simplified.

The Tri-Labs ASC program set an aggressive cost-reduction goal for its program, and Appro’s SU design permitted the economies of scale — in both components and systems — to achieve ASC’s objective. Overall, the ASC program estimated a significant reduction of its total cost of ownership (TCO), thanks to Appro’s flexible, scalable and reliable Linux clusters in conjunction with outstanding project management and customer installation services. In the end, their overall TCO story was very compelling.

Appro has now leveraged the idea of the scalable unit for all current and new supercomputing cluster solutions, including the Xtreme-X supercomputer and its scalable software management package. We refer to the entire SU-based architecture as the Scalable Supercomputing Cluster Architecture.

What kind of customers need or want Appro’s Scalable Supercomputing Cluster Architecture?

DK: Our customers are scientists, engineers and financial analysts who perform highly computational or data-intensive tasks and who need supercomputing resources to power their scientific research discovery, data modeling or seismic research. Therefore, our customers require high performance, high reliability and excellent system management for a wide range of HPC applications. Specific applications include computational fluid dynamics, computer-aided engineering simulations, petroleum exploration and production, scientific visualization for oil discovery and recovery, and research in seismic, defense and classified projects.

The goal of greater performance at a lower cost has thrust clusters to the forefront of HPC. Still, HPC users are demanding ever-better reliability, availability, manageability, compatibility and power efficiency. Responding to these demands, Appro’s Xtreme-X Supercomputer heralded a brand new architecture for HPC: the Scalable Supercomputing Cluster Architecture. This architecture groups clusters together using the scalable unit design to make a unified, fully integrated system that can be provisioned and managed as a stand-alone supercomputer. We feel that the Appro Xtreme-X Supercomputer and our Scalable Supercomputing Cluster Architecture address our customers’ needs for highly computational and data-intensive supercomputing.

Explain some more about the Appro Xtreme-X Supercomputer series. Why is this product line important?

DK: The Xtreme-X Supercomputer Series shows that Appro has paid close attention to current HPC user “pain points” and the evolving requirements of the technical server market. The Appro Xtreme-X series is ideal to scale out datacenters from 2.7 Teraflops to over 1,000+  Teraflops of computing power. It delivers significantly reduced TCO, energy-efficient (green) architecture and a complete lights-out management system to meet the demands of scalable performance and high-availability features.

The Xtreme-X Supercomputer product line is important because it is aligned with the demands of the HPC market today, and it will scale to continue meeting evolving needs in the future. HPC has now moved well beyond its origins in large government and university research sites. Today, HPC is indispensable for large commercial firms and is quickly moving into small and medium-sized enterprises (SMEs) across a broad range of vertical markets. The rise of open and standards-based HPC clusters has made this transformational technology accessible to even small firms and workgroups. Also, for the first time we see information on HPC market trends, data and examples showing how HPC is being used today in sites ranging from leading government and university centers to business and industrial firms, to produce faster, superior innovation and solutions. HPC is a growing market that business of all sizes will adopt so they can compete and survive.

Today, the Appro Xtreme-X Supercomputer Series offers the Xtreme-X1 model, based on dual-socket, quad-core Intel® Xeon® processors, and the Xtreme-X2 model, which supports quad-core AMD Opteron™ processors. Both models are based on Appro’s Scalable Supercomputing Cluster Architecture.

Appro intends to continue providing the highly reliable supercomputing platform and added-value features required by HPC customers, while accelerating businesses’ competitive advantage and reducing their TCO.

Appro recently announced the first demonstration of 40 Gb/s InfiniBand supercomputing clusters. What does this mean for the HPC market, and what kinds of customers will benefit?

DK: 40 Gb/s Infiniband supercomputing clusters address a critical need for faster, low-latency and larger bandwidth for large-scale deployments, with the ability to use any standard PCI Express Adapter available today.

Today, server and storage systems are deploying multiple multi-core processors. In these systems, overall platform efficiency and CPU and memory utilization depend increasingly on interconnect bandwidth and latency. For optimal performance, platforms with several multi-core processors can require interconnect bandwidths of more than 10 Gb/s or even 20 Gb/s. Supercomputers that can deliver 40 Gb/s bandwidth and lower latency – helping to ensure that no CPU cycles are wasted due to interconnect bottlenecks — will deliver unparalleled performance for the most demanding applications. They will take HPC to an even higher level of performance.

All HPC applications will benefit from this technology, including bioscience and drug research, data mining, digital rendering, electronic design automation, fluid dynamics and weather analysis. These applications require the highest throughput, to support the I/O requirements of the multiple processes they use for accessing large datasets to compute and store results. All these HPC applications are ideal for 40 Gbs supercomputing clusters.

What tools and solutions does Appro provide to aid in the management of these complex HPC systems?

DK: We offer the Appro Cluster Engine™ (ACE) management software, which features a complete lights-out management solution. This software suite provides a Web-based management interface that is easy to use, making it possible to control the Appro Xtreme-X supercomputer from any location.

The management modules include Network Management, Server Management, Cluster Management and Storage Management. In addition, the ACE software supports diskless configuration and network failover to achieve maximum reliability, performance and high availability. It supports root file systems offering instant provisioning for rapid, standard Linux installs on large diskless systems, allowing them to boot 64 to 6,400 blades at the same time. The Appro Cluster Engine management software offers reliable, available and serviceable (RAS) features in a total software management package.

Appro recently decided to partner with NEC. What does this partnership mean for the two companies and for HPC customers worldwide?

DK: Appro and NEC have been negotiating a partnership agreement for a while. NEC addresses the needs of a wide variety of HPC organizations with its dedicated Europe, Middle East and Africa (EMEA) HPC channel division. For our part, Appro would like to expand our channel geographically. With NEC’s strong technology base in EMEA and Appro’s cluster deployment successes in the HPC market, this partnership provides a sustainable competitive advantage enabling both companies to take a greater share of this growing market segment.

Appro’s Xtreme-X Supercomputer and Appro Cluster Engine management software are now being added and branded as part of NEC’s HPC solution offering in EMEA. Appro will continue to focus its sales efforts in the U.S., but having NEC as a strategic partner marks a breakthrough for Appro’s supercomputers entry into the EMEA HPC market. The partnership will enable Appro and NEC to work together toward a common goal, focusing on reducing complexity of technology integration when deploying and managing integrated solutions – and lowering customers’ TCO.

Are you concerned that the NEC partnership, in which Appro’s core products are marketed as part of NEC’s HPC offerings in the EMEA region, will dilute Appro’s brand awareness or strength?

DK: Instead of viewing this relationship as a dilution of Appro’s brand — which is very strong in the U.S., where we’ll continue to market products under the Appro name — I see this partnership as an extension of Appro’s reach into EMEA HPC markets. From NEC’s point of view, they now can offer flexibility and choice for customers to select high-value supercomputing solutions with reduced TCO.

As a side note, we are happy to announce that Appro has provided a benchmark cluster that is now up and running at the NEC facility in Houston, where most of the benchmarks will take place for EMEA customer opportunities. The arrangement is very convenient for both parties, since Appro also has a sales and service office in Houston. The teams are taking advantage of their proximity to each other to work closely together.

What are some new HPC technology or product directions we should be anticipating from Appro?

DK: The benefits of HPC solutions include higher performance and scalability — and to a lesser extent, investment protection and simplified administration. Appro will continue to deliver cutting-edge HPC technology and solutions to our customers.

We believe that the Xtreme-X Supercomputer will have a huge impact on HPC markets because it addresses most of our customers’ key pain points. Appro is also looking to offer a new line of server solutions early next year that will support the latest x86 processors as well as GPU computing. We are proud of our up-to-date Xtreme-X Supercomputer deliverables and accomplishments this year, and we look forward to further enhancing our overall high-performance computing product offerings and extending our value proposition.

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