SGI Sticks to Its Roadmap, But It’s the End of the Line for the Itanium Supers

By Michael Feldman

July 22, 2009

It’s been just a little over two months since Rackable Systems acquired SGI and merged the two product lines. From all appearances, it has been a fairly smooth transition. A lot of that has to do with the fact that the new company has left both sets of product offerings intact, although the future “Ultraviolet” system will change that somewhat. I’ll get to that plot line in a moment.

According to Geoffrey Noer, senior director of product marketing at SGI, they “have not discontinued any of the products and have no immediate plans to do so.” The product continuity is a reflection of the company’s intent to combine Rackable’s established customers in the Internet/cloud computing space with SGI’s strength in the high performance computing market. Even before the merger deal, Rackable was looking at the HPC technical and federal government markets as a way to grow the business. Now that they have the SGI Altix products in tow, they get to those markets by default.

Except for the adoption of SGI as the new company name, most of the product branding will remain intact too, with just some minor tweaks. The Rackable name, itself, will live on in the company’s rack-mount server line. This consists of the half depth x86 servers, which can be mounted back to back, plus the standard depth servers. One change is that the SGI Altix XE line has been folded into Rackable server portfolio. The rack-mount machines represent the company’s most flexible systems, configuration-wise.

The CloudRack C2 offerings are also from the Rackable legacy, and as you might guess are aimed at cloud computing and other large-scale datacenter type applications. The CloudRack tray design is sort of a cross between a rack-mount and blade architecture. To boost energy efficiency — a big Rackable focus, even before it became fashionable — DC power as well as fan cooling is supplied at the rack level rather than inside each server. Noer says the system is 93 percent efficient from the plug to the server.

The CloudRack’s real claim to fame is its adaptability for hot datacenter environments, with some configurations rated for air temperatures as high as 40C (104F). One of the latest trends in datacenters recently has been to let temperatures rise in order to help cut air conditioning costs. The problem is that the system architecture needs to take this into account, otherwise the power you save in AC just gets re-spent by the fans. The CloudRack approach is to use large external fans and place the hard drives, which tend to be one of the more heat-sensitive components, at the front of the rack where it gets the coolest air. The CPUs are more heat-tolerant, so they go toward the back. With this design, a kilowatt of fans can cool the 80 or more servers, according to Noer. The idea is to be able to use ambient outside air for much of the year, and in cool summer climates, it may be possible to do away with AC altogether.

The ICE Cube is SGI’s containerized datacenter offering. And despite the ICE designation, it was developed under the Rackable regime. Containerized datacenter solutions have been all the rage lately, with every major system vendors offering some version of this. Noer says they have SGI customers who are interested in deploying Altix 4700 shared memory HPC machines in containers, and they’re looking into how this might be accomplished.

In fact, even though the product lines mentioned are aimed at the scaled-out datacenter, SGI believes they can also be sold into HPC accounts as well. The rack-mount servers have had some success in this space even before the merger, and Noer says they have future plans for HPC-centric support in their CloudRack line too.

The Altix ICE 8200 and Altix 4700 product lines from the legacy SGI business will remain in place at the heart of the HPC business. The ICE 8200 is based on Intel Xeon-based blades, while the 4700 is SGI’s latest shared memory system powered by the Itanium processor and the proprietary NUMAlink system interconnect. The shared memory machine enables up to 512 processors (1024 cores) and 120 TB of memory to be under the control of a single image of the operating system.

The Itanium-based machines have been popular with a few high-end HPC users looking for a shared memory programming model. But even in the TOP500 list, there are only three SGI Altix shared memory systems, one of which is NASA’s Columbia machine, which is based on the older 3700 architecture. That compares to 15 SGI ICE 8200 systems on the TOP500, including the number four-ranked Pleiades supercomputer at NASA. The decline of Itanium’s fortunes in high performance computing has paralleled the rapid rise of x86 performance and its support for cache coherent NUMA. The fact that Intel was chronically late on introducing new versions of Itanium didn’t help the processor’s cause.

Well, apparently SGI is moving on as well. Noer says they will continue to sell the 4700, but the next generation shared memory system will be based on the new Ultraviolet architecture. That design will use Intel Nehalem EX chips along with the next generation NUMAlink interconnect. Presumably this means the future “Tukwila” quad-core Itanium chips will never find a home at SGI*. Although the 4700 line will continue to be offered for some period of time, the idea is to eventually migrate all the current users to the new architecture. “The intention is that Ultraviolet is the future of the shared memory systems line,” says Noer.

According to him, more information about Ultraviolet will be publicly revealed later this year (although SGI customers have been privy to some of the details for awhile now). I would expect to see the introduction of the first new machines coincident with Intel’s release of the Nehalem EX processors, which probably means early 2010.

Bringing the entire server product portfolio under x86 won’t be a panacea for the new SGI, but it’s a step in the right direction. The competition in the HPC system arena is still rather formidable despite the loss of Linux Networx and SiCortex. If SGI manages to find a comfort zone below the tier one vendors by being more nimble and willing to go after both large and small accounts, it should be able to ride out the tough economy until better times return.

[*UPDATE:  According to the latest blog by CEO Mark Barrenechea, SGI’s support for Itanium may extend much farther than what is described here. Mark at least implies that both Xeon and Itanium will be part of the shared memory Ultraviolet systems in the future and he portrays SGI’s committment to Itanium as “100%.” I suggest you take a look at the full blog entry to form your own opinion.]

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