IBM Dials Up Density for HPC and Hyperscale

By Nicole Hemsoth

September 11, 2013

IBM is casting an ever-widening net to cover a broader range of workloads starting with today’s announcement of its NextScale system, which is designed to use a stripped-down x86 lure to reel in everything from the cloudy heights to high performance computing.

That lure is surprisingly simple–and all by design. Server buyers on all sides of the compute spectrum are being pulled into the swift current of stripped-down, purpose-driven and cost-conscious hardware. IBM has remained one of the few that continued to swim upstream of this raw-box approach–but it’s catching up–and throwing some performance awareness into the mix.

Adding to their x86 drive, the star of NextScale is the Ivy Bridge processor (one or two) which are nestled within a half-width NextScale nX360 server. These snap snugly into the host n1200 enclosure (a 6U, 12-bay chassis) that can double up to host 84 of these pared-down boxes—or 2,016 cores–in a standard 19-inchrack.

And “standard rack” is the key word here, at least for those who bit the iDataPlex line but were scared off by the strange manipulations of space and time necessary to maneuver around its (useful but) non-standard design. In other words, IBM has taken a “best of all worlds” approach with the NextScale system, pulling the MVP features from iDataPlex and Flex alike to create something that might be able to go head-to-head with the wave of hyperscale (and hypercheap) solutions that are flooding the market.

This new server strain puts IBM in a much larger petri dish with competitive offerings from HP (namely the SL6500 series and the half-width SL390S) as well as similar lines from Dell (the C8000 series, in particular). But the difference here, says System X Product Manager, Gaurav Chaudhry , is that they’re able to offer integration with some of IBM’s key initiatives and products, including their recently acquired workload management tools from Platform Computing, full GPFS support and for the cloudy side, ready-made APIs and binaries for xCAT which, while free to begin with, is being actively supported by IBM.

Dubbed the “economical addition to the System X family” IBM says that this approach offers the density, performance and flexibility to support the diverse targeted workloads. For the HPC crowd, there are certainly some features worth noting—but generally speaking, this is a pared-down approach that lets users build what they need at a price point that’s relative to other bare-bones boxes from competitors. But there are still some things that are off in the future—enough so that we might not see many of these finding their way into the Top 500 before next ISC.

The NextScale announcement is really about possibility and the future, at least for HPC buyers. So far, this is an Intel-only offering, with support for GPUs and Xeon Phi coming in early 2014. While there were no timeframes stated for other additions, including Power, ARM or other processors, it stands to reason that IBM wouldn’t want to be left behind as others march to the beat of customer demand.

With that said, to snap in spicy elements like GPUs or Phi means a need for more sophisticated cooling. One notable missing element with NextScale is the direct water cooling. The version announced today is air-cooled with the possibility of passive water cooling. Chaudhry says that direct water is coming right in line with other acceleration/co-processor capabilities.

That omission aside, there are some other notable elements that will appeal to the HPC set. For instance, IBM will offer well-rounded support for Infiniband, including FDR and QDR. There is no integrated I/O or switching, no chassis-level management, and the attractive part of iDataPlex, namely its front access capabilities to almost all components, was carried over.

For the cloud and general datacenter users, there will be two standard gigabit Ethernet ports, although they’ll be glad to sell additional capability to add 10GbE as an option as well. While HPC is all about the IOPS, latency and general performance, Chadhury says that the cloud customers simply want to get up and running as soon as possible—it’s all about time to delivery, he says

As IBM’s David Watts noted of the new nx360 server addition:

The IBM NeXtScale nx360 M4 compute node contains only the essential components in the base architecture…

  • 1 or 2 Intel Xeon E5-2600 v2 series processors
  • Up to 8 ECC DIMMs operating up to 1866 MHz providing a total memory capacity of up to 128 GB
  • 2 on-board 1Gb Ethernet ports
  • Integrated management with the IMM2 supporting IPMI and UEFI industry standards
  • PCI Express 3.0 I/O expansion capabilities through a 16x riser cage
  • Additional mezzanine slot for FDR InfiniBand or 10Gb Ethernet without consuming a PCIe slot

IBM  took their cues from the positive responses around offering front accessibility (as with the iDataPlex systems) and carried that over to NextScale. Chaudhry freely admits that while iDataPlex was made for its own configuration and offered little flexibility, the idea with NextScale was to “keep all the things we liked and get rid of other things, like the full-width server versus this 8.5 inch wide but deep (versus shallow) approach.”

On that “best of all worlds” note touched on earlier, Chaudhry highlights the difference between the Flex versus NextScalelines s in the increased ability to go as “vanity-free” as one wishes. “So say if a customer isn’t looking for the integrated switching built into the chassis, they have the flexibility to pick their own,”

The key concepts behind this launch are around flexibility, simplicity and scale—in short, a move away from the tricky design and implementation details of their iDataPlex but with more room to grow than Flex might offer for some users. IBM Product Marketing Manager for the System X line, Gaurav Chaudhry says that this isn’t the immediate end of the line for iDataPlex. It simply marks an evolution toward flexible systems that can meet the low latency, high performance, and I/O demands of HPC while remaining lightweight and simple enough for cloud users to hop into without a great deal of effort. The company still has a number of iDataPlex systems to support, which will continue for at least 18 months, but the hyperscale, low-cost NextScale is the real battlefield for IBM’s push—and the shove will come down to price.

As a side note, it seems a shame that the Flex line was named “Flex” versus NextScale—seems the true meaning of these are flip-flopped, with the Flexes emphasizing scalability and NextScale pushing flexibility.

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