E4 Computer Engineering Unveils New ARM-GPU Clusters

By Nicole Hemsoth

November 13, 2012

This week at SC12, Italian cluster maker E4 Computer Engineering, launched a new series heterogeneous clusters, which pair an NVIDIA’s ARM+GPU Tegra3 with a discrete Quadro GPU. We asked E4’s Simone Tinti, who leads the HPC team at E4, to describe the new systems and talk about the advantage they offer to high performance computing users.

HPCwire: What does E4 Computer Engineering do, and who are your primary customers today?

Simone Tinti: E4 Computer Engineering new web site went online just before SC12 with a renewed look and a lot of contents) designs and manufactures high  performance computing and storage systems. We have a vast portfolio of solutions ranging from technical workstation to complete datacenter. E4 is also very active in providing storage solutions.

E4 Computer Engineering is currently operating in Europe, although CARMA solutions will be available worldwide. Our primary customer base is academia and research. For the past 7 years we have been one of the major supplier of CERN. At present we have more than 11,000 cores active in the CERN  datacenter –1700 computing nodes — and more than 20 petabytes of storage.

Other relevant customers are ETHZ, EPFL, INFN, SISSA, ICTP,  Novartis, Merck, NATO and many more.
 
CARMA solutions are the results a E4 and SECO partnership. SECO is a European  designer  and manufacturer  of  high  integrated  board computers  and  systems  for  embedded applications. Founded in 1979 in Italy, SECO attention is focused on developing   innovative products with high performances efficiency, low power    consumption and increased functionality, offering in the meantime a short time-to-market. SECO has designed and manufactured the CARMA Devkit
 
HPCwire:  In a nutshell, could you briefly describe the three ARM-based solutions you are introducing and the application domains they are targeted at?
 
Tinti: We are introducing two platforms able to host carrier boards for the Qseven modules.

The CARMA microcluster, a 5U chassis, contains up to 8 CARMA blades  — by default they have one NVIDIA Tegra 3 plus one NVIDIA Quadro 1000 — and one x86-based  management node.

The CARMA cluster is a 3U chassis, containing up to 12 CARMA2 blades. These are high density blades, each one containing two CARMA blades for a total of two ARM CPUs plus two Quadro 1000 GPUs or 12 DARMA blades with four ARM CPUs per blade. You can, of course, mix DARMA and CARMA2 blades in the same chassis.

All blades (CARMA, CARMA2 , DARMA) are compliant with the Qseven standard so  the motherboard  — or to be more accurate the carrier board — contains only I/O devices like network, HDMI, SATA and so on.

The CPU component  — actually a SoC containing CPU, RAM, Flash memory, graphics adapter — may be selected from broad range of x86 architecture http://www.seco.com/it/itemlist/qseven/. Qseven technology  provides the highest flexibility while reducing engineering costs. In fact you can swap the CPU architecture re-using  the same carrier board without any modification.  SECO is a co-founder of Qseven consortium.

HPCwire: Regarding HPC applications, what are the advantages of the CARMA solutions compared to a more traditional x86/Tesla GPU cluster?  What niche are you filling with these ARM-based solutions?

Tinti: CARMA solutions provides a low power platform, ideal for applications that relies mainly on GPU computing power. With typical GPU computing systems you need a platform in the range of 250 to 350 watt, for example, a dual-Intel Xeon E5 or dual-AMD Opteron 6200 or 6300 in order to have your GPUs up and running. This is fine as long as a relevant part of the computation relies on CPUs, otherwise this is simply a waste a of power.

Usually GPUs are claimed to have a gigaflops per watt ratio of 3 (higher that the ~2 you can achieve on a BlueGene/Q systems, this is true only if you consider the GPU devices — around 200 watts for  600 peak gigaflops. When you consider the whole platform this ratio drops down 1.2 /0.9 not far from a pure CPU systems as documented in the Green500.

With CARMA blades you need only 10 watts for the CPU, RAM, and flash drive; therefore most of the power — 45 watts — is dedicated to GPUs.

The “must” for exploiting CARMA solution is to have an application that is strongly focused on CUDA for CARMA and CARMA2 blades, or on big data/cloud for DARMA blades.

HPCwire:  What is the advantage of pairing a heterogeneous ARM plus GPU Tegra 3 with a discrete Quadro GPU?  What is the intended programming model for such an arrangement?

Tinti: With the current generation of ARM CPUs, you cannot address algorithms based on floating point arithmetic, therefore most scientific applications are excluded. NVIDIA Quadro GPU broadens the range of applications that can be addressed and gives a huge boost to performance.  CARMA is the one and the only platform available on the market that combines ARM’s low power CPUs with powerful NVIDIA GPUs.

HPCwire: Will developers with existing CUDA applications, run on an x86-Tesla set-up, be able to port their codes to the CARMA platform?

Yes, we provide a pre-configured cross compiling environment for ARM, CUDA, and MPI that makes this process very easy. Support services are also available. We are currently porting some applications, which will be disclosed soon. The systems come with the NVIDIA SDK and ORNL’s SHOC benchmark suite.

HPCwire: Are these clusters intended for production environments?

Tinti: The CARMA cluster is designed to be used in production environment, and provides a robust platform for a wide range of application: HPC, big data, and cloud.

The CARMA microcluster is designed to be a perfect development platform. It’s very quiet and can also be placed beside a desk. It could be used to create, but not for critical environments since a redundant power supply or remote management feature, like IPMI, are not available yet. Based upon the feedback we will receive at SC we will eventually release a more robust version of CARMA microcluster.

HPCwire:  Do you have customers with installed systems, or in the pipeline, for any of  the CARMA or DARMA systems?  What geographies do you intend to serve?

Tinti: More than 2000 CARMA dev kit has been sold so far to the most relevant research centers around the world. A lot of industries in different market such as animation, oil & gas, microelectronics, telecommunications, defense, and manufacturing have adopted it as a development platform. Unfortunately we cannot disclose the name, most of them are developing innovative applications and prefer to keep their privacy right now. Most of these customers are of course waiting for a platform ready for production, like the CARMA series.

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!

ARM, Fujitsu Targeting Open-source Software for Power Efficiency in 2-nm Chip

July 19, 2024

Fujitsu and ARM are relying on open-source software to bring power efficiency to an air-cooled supercomputing chip that will ship in 2027. Monaka chip, which will be made using the 2-nanometer process, is based on the Read more…

SCALEing the CUDA Castle

July 18, 2024

In a previous article, HPCwire has reported on a way in which AMD can get across the CUDA moat that protects the Nvidia CUDA castle (at least for PyTorch AI projects.). Other tools have joined the CUDA castle siege. AMD Read more…

Quantum Watchers – Terrific Interview with Caltech’s John Preskill by CERN

July 17, 2024

In case you missed it, there's a fascinating interview with John Preskill, the prominent Caltech physicist and pioneering quantum computing researcher that was recently posted by CERN’s department of experimental physi Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to communities across the globe. As climate change is warming ocea Read more…

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical claims. A paper published on July 10 by researchers in the U. Read more…

Belt-Tightening in Store for Most Federal FY25 Science Budets

July 15, 2024

If it’s summer, it’s federal budgeting time, not to mention an election year as well. There’s an excellent summary of the curent state of FY25 efforts reported in AIP’s policy FYI: Science Policy News. Belt-tight Read more…

SCALEing the CUDA Castle

July 18, 2024

In a previous article, HPCwire has reported on a way in which AMD can get across the CUDA moat that protects the Nvidia CUDA castle (at least for PyTorch AI pro Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to com Read more…

Shutterstock 1886124835

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical c Read more…

Shutterstock 2203611339

NSF Issues Next Solicitation and More Detail on National Quantum Virtual Laboratory

July 10, 2024

After percolating for roughly a year, NSF has issued the next solicitation for the National Quantum Virtual Lab program — this one focused on design and imple Read more…

NCSA’s SEAS Team Keeps APACE of AlphaFold2

July 9, 2024

High-performance computing (HPC) can often be challenging for researchers to use because it requires expertise in working with large datasets, scaling the softw Read more…

Anders Jensen on Europe’s Plan for AI-optimized Supercomputers, Welcoming the UK, and More

July 8, 2024

The recent ISC24 conference in Hamburg showcased LUMI and other leadership-class supercomputers co-funded by the EuroHPC Joint Undertaking (JU), including three Read more…

Generative AI to Account for 1.5% of World’s Power Consumption by 2029

July 8, 2024

Generative AI will take on a larger chunk of the world's power consumption to keep up with the hefty hardware requirements to run applications. "AI chips repres Read more…

US Senators Propose $32 Billion in Annual AI Spending, but Critics Remain Unconvinced

July 5, 2024

Senate leader, Chuck Schumer, and three colleagues want the US government to spend at least $32 billion annually by 2026 for non-defense related AI systems.  T Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then Read more…

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_1687123447

Nvidia Economics: Make $5-$7 for Every $1 Spent on GPUs

June 30, 2024

Nvidia is saying that companies could make $5 to $7 for every $1 invested in GPUs over a four-year period. Customers are investing billions in new Nvidia hardwa Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

AMD Clears Up Messy GPU Roadmap, Upgrades Chips Annually

June 3, 2024

In the world of AI, there's a desperate search for an alternative to Nvidia's GPUs, and AMD is stepping up to the plate. AMD detailed its updated GPU roadmap, w Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very Read more…

Leading Solution Providers

Contributors

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l 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…

IonQ Plots Path to Commercial (Quantum) Advantage

July 2, 2024

IonQ, the trapped ion quantum computing specialist, delivered a progress report last week firming up 2024/25 product goals and reviewing its technology roadmap. 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…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh 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…

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…

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…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire