Cray Unveils Personal Supercomputer

By John E. West

September 16, 2008

With few exceptions in its 30-year history, Cray has focused exclusively on the high end of the HPC market. This week the company announced a product aimed at a new space: deskside HPC. Ian Miller, Cray’s senior vice president of sales and marketing, talked with HPCwire about the company’s new CX1 personal supercomputer.

HPCwire: You’ve just announced a new personal supercomputing product called the Cray CX1. What is it?

Miller: The CX1 is Cray’s new personal supercomputer. The unit is small — it’s meant to fit beside a desk — and it can be plugged into a wall socket on standard office power. Users can configure it in a variety of ways to suit their specific needs, but fully outfitted for computation, it will hold up to eight nodes, each with two dual- or quad-core Intel Xeon processors, and up to 64 GB of memory per node. The machine is very versatile, with configuration options that allow users to mix and match storage, compute and visualization capability in the same box. The system can be configured with up to 4 TB of storage, and comes with Gigabit Ethernet or InfiniBand networking.

Customers will be able to pick the operating system that best matches their applications and their expertise. The CX1 comes with either RedHat Linux or, for the first time ever on a Cray product, Windows HPC Server 2008.

HPCwire: This is quite a departure for Cray. Who is this computer for?

Miller: We’re launching the CX1 with the tag line “we take supercomputing personally,” and that really reflects the whole philosophy we’ve adopted while developing this machine. The CX1 provides supercomputing where you need it. It’s designed to run on standard 110 volt office power, and doesn’t require additional cooling. It can sit under your desk, right where you work.

It’s built for individual team and workgroup deployment, providing up to 64 cores (in a single chassis) of processing power that will enable users to run small and medium-sized jobs without the need to sit in batch queues and compete with other users for resources. And it can be configured to provide the particular mix of capabilities users need. While the CX1 holds up to eight compute nodes, customers can buy just the right number at first and then expand as their needs grow, and they can supplement compute nodes with storage and visualization nodes to make sure they have the right balance of capabilities that fit their needs.

The CX1 will fit well in workgroups and departments that need their own parallel processing capabilities, in business that need high performance computing for engineering or business intelligence, or even for developers in traditional high end supercomputing facilities who need a responsive, affordable environment for developing and debugging HPC software.

HPCwire: Why does this product make sense at this particular time? What hole in the market is it filling?

Miller: The CX1 is about helping customers overcome the obstacles to HPC adoption. Although there have been many recent accounts of the barriers to entry for customers new to HPC, the “Reveal” and “Reflect” reports from the US Council on Competitiveness are great recent examples. These reports identify three significant barriers: lack of application software, lack of sufficient talent, and cost constraints. We feel that we’ve gone a long way toward addressing all three of these with the CX1.

First, the CX1 uses standard operating systems. Both RedHat Linux and the Windows family of operating systems standard in offices and research groups around the world, and both are widely supported by software vendors. And to make sure that the available pool of application software continues to grow, Cray is building partnerships with a number of the best known ISVs in segments like manufacturing, life sciences, financial services, and digital media.

Second, we’ve tried to lower the bar in terms of the talent required to deploy one of these at a customer site. In terms of infrastructure, the CX1 will just drop into a standard office environment. The operating system and tools will be very familiar to Linux or Windows users, and Cray is offering a 3-year warranty and onsite support for those times that something out of the ordinary happens and customers need an expert to come look at the system.

Finally, the CX1 starts at just $25,000, with fully configured systems reaching the $80,000 range. The system is very affordable in terms of both the initial capital investment and the lowered total cost of ownership customers will see from ease of management and standard office power and cooling requirements.

HPCwire: At its heart, the CX1 is still a cluster, and clusters are notoriously hard to configure, purchase and deploy. Can you talk a little more about how you’ve tried to put the CX1 within reach of non-specialists?

Miller: This is a system designed for non-specialists all the way from purchase to submitting the first job. The CX1 can be purchased on the Web using an online configuration tool that is oriented to the needs of customers in specific industries. The tool will help customers find the system configuration (in terms of dual or quad-core, amount of memory, and so on) that best meets their specific needs for the kinds of application codes that are typical of their industry. Of course, more advanced users will be able to specify precisely what they want.

Each system will ship in a single box, with color-coded cabling, and software pre-installed for ease of setup. System configuration is driven by wizards and GUI interfaces that make setup straightforward, even for first-time cluster owners. We are also building out a Web community where users will be able to take advantage of the community of CX1 users and Cray specialists to offer support and advice to one another.

Although this is a different kind of system, it’s still a Cray, and we want to make sure that CX1 customers have the kind of experience Cray customers have come to expect from us. Each CX1 will come with a 3-year warranty and next business day onsite support, standard.

HPCwire: Seeing Windows on this Cray product is a big surprise. Can you talk about what motivated this decision?

Miller: We believe that there are many workstation users today who are used to working in a Windows environment and find the thought of moving to a more powerful platform like an HPC cluster and the challenge of learning a new operating system daunting. By offering an operating system that they are familiar with, we believe the barriers to adoption are significantly lowered. Our approach and goals for the Cray CX1 are in alignment with Microsoft’s — to enable more people to move to high performance computing and the productivity improvements that will result.

HPCwire: What are the specifics of available processors, storage, and graphics options?

Miller: The Cray CX1 is highly versatile with a variety of blades that can be mixed and matched to customize the system to meet varying user requirements. It is designed with eight nodes in a chassis with eight slots for blades. The CC48 is a single-socket and the CC54 is a dual-socket blade — both with dual- or quad-core Intel Xeon processors and 8 DIMM slots per blade. A single chassis supports up to 16 Intel Xeon dual-core or quad-core processors — for a maximum of 64 cores. In addition, each node can accommodate up to 64GB of memory.

This flexibility extends beyond just compute blade options — the CX1 can also be configured with one or more visualization blades — something not even available in any other system of this size. The CV5401 visualization blade features NVIDIA Quadro FX graphics cards as well as an optional NVIDIA GPU.

Additionally, there are two different storage blade options, the CS5404 and CS5408, which combine Intel Xeon’s for computing with four or eight 2.5 inch Serial Attached SCSI disk drives for storage. Configuring a system for maximum I/O results in up to 4 terabytes of storage in a single Cray CX1 chassis.

HPCwire: Can I put more than one CX1 together in a system?

Miller: Absolutely. You can actually cluster three CX1’s together — for up to 192 cores of compute — without an external switch, and you can grow your cluster beyond that with some additional switch gear. And because we are committed to an “ease of everything” experience for non-traditional HPC customers, we are building this to be an Intel Cluster Ready system.

HPCwire: Cray is a company that has traditionally focused on delivering solutions for the very high end of this space. The CX1 is at the opposite end of the HPC market, and will require a totally different approach to both sales and support. How is the company going to manage such disparate product models?

Miller: We have approached the Cray CX1 differently, with a lower-touch sales process and tighter relationships with strategic partners, like Microsoft and Intel. They CX1 can be ordered easily from our Web site or through a dedicated inside sales team that we’ve established. Cray has also established certified service providers around the world to provide the quality service expected from Cray. So the introduction of the Cray CX1 is not distracting us from our core, high-end business.

HPCwire: Are we seeing a new Cray?

Miller: Not a new Cray, but a new aspect to Cray. What we are seeing is Cray’s response to the HPC needs of a new market. This is a very exciting time to be in high performance computing. The move to multicore processors is leading to an explosion of developers facing the same problems, and using the same tools and techniques that have been the sole province of the supercomputing community for 50 years. For perhaps the first time in modern history, the flow of innovation in high performance computing is going to be driven by the millions of programmers focused on commodity application development. This will mean a proliferation of new ideas, new tools, and new techniques with the potential to completely reshape high end computation. With decades of experience in supercomputing, Cray is ideally positioned to help lead this transformation by providing tools that will help open up the promise of high performance computing to a broader community of users than ever before.

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues 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 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…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion 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…

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