Even Better than a ‘Supercomputer-in-a-Box’

By Gary Keen, Engineer, Silicon Mechanics

March 15, 2021

Before we get started, can everyone reading this article agree that large compute clusters, supercomputers, HPC systems, or whatever else we choose to call them, are complex? Ok, good! Because, while “out of the box” solutions make things somewhat easier, it’s an oversimplification to say that’s all you need – and you end up missing a few key points about the value of complexity to the end user and their organization.

Many, many individual pieces need to work together to create a cluster and all of them can have a huge impact on the system’s overall performance. From a system architects’ perspective, each component is a decision that must be made, a choice that either improves your performance for a particular workload or hurts it. These decisions can weigh heavy on architects, but often weigh even heavier on the IT infrastructure managers responsible for their company’s budget and resources.

Bad choices in system design have a huge impact to a client organization’s capabilities and, ultimately, revenue. The problem is that it’s hard to know what to tell the system architect about which choices you prefer.  Yes, you could design the system yourself, but who has time to bone up on all the issues you need to consider and learn all the technologies out there that might work? Every choice could be a mistake – and your team could suffer for it for years.

From that point-of-view, it’s easy to see why the ‘supercomputer-in-a-box’ idea is appealing to most people on some level. A major technology provider with tons of experience, resources, and engineering brainpower figures out the best possible system for HPC, AI, or data analytics and you buy it.

Maybe it won’t be perfectly tuned to your workload, but at least you know what you can expect, and you don’t have to worry over the little decisions. It’s comfortable, predictable, and relatively safe. It’s also expensive, restrictive, and commonplace. Commonplace is important because, if everyone buys the same solution, no one has a competitive advantage. And who doesn’t want a competitive advantage?

Instead, you could get the best of both worlds by working with a solution provider with experience in both the latest technology but also a solid foundation in proven platforms and have this person design something unique to your needs.

The result is that you get a one-of-a-kind solution tuned to your workload that also stretches your budget further than the preconfigured solution could. What’s even better is that, today, that knowledgeable system architect you’ve hired will be aware of how you can take advantage of the best parts of the preconfigured solution.

Specifically, your system architect can give you the predictability, the simplicity, and the safety of that pre-configured supercomputer-in-a-box you keep seeing advertised everywhere and combine them with the best elements of a custom solution, namely lower costs, fine-tuned performance, and competitive advantage.

How? Building blocks.

Let me explain. System designers at reputable firms will have access to all the individual pieces that make up these pre-configured solutions. For instance, my team builds custom AI solutions based on the NVIDIA HGXTM server platform, which features the A100 GPU. When optimized to a specific workload, these custom solutions provide unmatched performance, and they can be tailored to be easy to use, seamlessly scalable, and more. Out-of-the-box solutions can’t give you that.

Generally, no two solutions are alike. This is great for customers that either need a unique solution for a unique problem or don’t have the budget for a pre-configured solution large enough to meet their needs.  These customers are willing to introduce variables to their system design in order to reach their goals. Not everyone is like that, and rightfully so, which is where the out-of-box options are most valuable.

Architects like our team at Silicon Mechanics want to reduce the number of variables in system designs to lower the perceived risk for our customers. We believe that building a strong solution for any workload requires balance between network, storage, and compute. So, we’re developing network, storage, and compute building blocks that are each unique, tested, and high-performance, but have their own, specific purpose in a larger system design.

Take storage for instance. A common trend nowadays is the faster the better, but that is only somewhat true. Not all workloads take advantage of that speed to the fullest, so using the fastest NVMe drives available, or even persistent memory devices, may be an unnecessary expense for you. Instead, we configure building blocks for slower flash storage or even *gasp* spinning disks.

We often suggest a tiered storage structure, which combines varying levels of high-bandwidth and high-capacity storage to provide top-tier performance at a better price. A pre-configured solution may only have the fastest, most expensive storage option available. The decision is no longer which drives to use or what form factor of a storage server to choose, but instead what performance level of storage does your workload need, and how much data will you have.

With the building blocks approach, the key is staying flexible. By configuring and testing building blocks at the storage, compute, or networking level, we gain some of the predictability of a pre-configured solution but can still tune the overall system to match your specific goals. This means as a customer, you can start small, investing only in what you need now, because you can always expand that system later. This is like the concept of rightsizing a cluster, you don’t always need a system to rank high on the Top500 list, you just need it to work for your team.

If you have a fully custom solution, it may be difficult to scale that design. Whereas a pre-configured solution may mean you have to start from scratch with the new version, or even worse, you are paying a premium for an old design that is no longer competitive with the market.

So as a solution designer who loves building unique solutions from the ground up and tinkering with components to get the most out of a solution, even I can endorse going with the middle ground. There are still great reasons to go with a fully custom solution and we are happy to build that with you. Similarly, the pre-configured solutions may be right for you and can absolutely be successful. It’s about finding a partner that you trust, knowing your options, leveraging your budget, and finding the solution that’s right for your team.

Right now, the Silicon Mechanics team is using this building blocks approach to design scalable, world-class AI and HPC solutions that feature the latest and greatest GPU architectures from NVIDIA. You can think of it as supercomputer performance without a supercomputer footprint.  If that interests you, I suggest checking out NVIDIA’s GPU Technology Conference (GTC) where our team will be giving several presentations about how to use building block technology and other best practices for industry-specific solutions.

To learn more about how Silicon Mechanics’ approach to system design, visit www.siliconmechanics.com

 

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Why HPC Storage Matters More Now Than Ever: Analyst Q&A

September 17, 2021

With soaring data volumes and insatiable computing driving nearly every facet of economic, social and scientific progress, data storage is seizing the spotlight. Hyperion Research analyst and noted storage expert Mark No Read more…

GigaIO Gets $14.7M in Series B Funding to Expand Its Composable Fabric Technology to Customers

September 16, 2021

Just before the COVID-19 pandemic began in March 2020, GigaIO introduced its Universal Composable Fabric technology, which allows enterprises to bring together any HPC and AI resources and integrate them with networking, Read more…

What’s New in HPC Research: Solar Power, ExaWorks, Optane & More

September 16, 2021

In this regular feature, HPCwire highlights newly published research in the high-performance computing community and related domains. From parallel programming to exascale to quantum computing, the details are here. Read more…

Cerebras Brings Its Wafer-Scale Engine AI System to the Cloud

September 16, 2021

Five months ago, when Cerebras Systems debuted its second-generation wafer-scale silicon system (CS-2), co-founder and CEO Andrew Feldman hinted of the company’s coming cloud plans, and now those plans have come to fruition. Today, Cerebras and Cirrascale Cloud Services are launching... Read more…

AI Hardware Summit: Panel on Memory Looks Forward

September 15, 2021

What will system memory look like in five years? Good question. While Monday's panel, Designing AI Super-Chips at the Speed of Memory, at the AI Hardware Summit, tackled several topics, the panelists also took a brief glimpse into the future. Unlike compute, storage and networking, which... Read more…

AWS Solution Channel

Supporting Climate Model Simulations to Accelerate Climate Science

The Amazon Sustainability Data Initiative (ASDI), AWS is donating cloud resources, technical support, and access to scalable infrastructure and fast networking providing high performance computing (HPC) solutions to support simulations of near-term climate using the National Center for Atmospheric Research (NCAR) Community Earth System Model Version 2 (CESM2) and its Whole Atmosphere Community Climate Model (WACCM). Read more…

ECMWF Opens Bologna Datacenter in Preparation for Atos Supercomputer

September 14, 2021

In January 2020, the European Centre for Medium-Range Weather Forecasts (ECMWF) – a juggernaut in the weather forecasting scene – signed a four-year, $89-million contract with European tech firm Atos to quintuple its supercomputing capacity. With the deal approaching the two-year mark, ECMWF... Read more…

Why HPC Storage Matters More Now Than Ever: Analyst Q&A

September 17, 2021

With soaring data volumes and insatiable computing driving nearly every facet of economic, social and scientific progress, data storage is seizing the spotlight Read more…

Cerebras Brings Its Wafer-Scale Engine AI System to the Cloud

September 16, 2021

Five months ago, when Cerebras Systems debuted its second-generation wafer-scale silicon system (CS-2), co-founder and CEO Andrew Feldman hinted of the company’s coming cloud plans, and now those plans have come to fruition. Today, Cerebras and Cirrascale Cloud Services are launching... Read more…

AI Hardware Summit: Panel on Memory Looks Forward

September 15, 2021

What will system memory look like in five years? Good question. While Monday's panel, Designing AI Super-Chips at the Speed of Memory, at the AI Hardware Summit, tackled several topics, the panelists also took a brief glimpse into the future. Unlike compute, storage and networking, which... Read more…

ECMWF Opens Bologna Datacenter in Preparation for Atos Supercomputer

September 14, 2021

In January 2020, the European Centre for Medium-Range Weather Forecasts (ECMWF) – a juggernaut in the weather forecasting scene – signed a four-year, $89-million contract with European tech firm Atos to quintuple its supercomputing capacity. With the deal approaching the two-year mark, ECMWF... Read more…

Quantum Computer Market Headed to $830M in 2024

September 13, 2021

What is one to make of the quantum computing market? Energized (lots of funding) but still chaotic and advancing in unpredictable ways (e.g. competing qubit tec Read more…

Amazon, NCAR, SilverLining Team for Unprecedented Cloud Climate Simulations

September 10, 2021

Earth’s climate is, to put it mildly, not in a good place. In the wake of a damning report from the Intergovernmental Panel on Climate Change (IPCC), scientis Read more…

After Roadblocks and Renewals, EuroHPC Targets a Bigger, Quantum Future

September 9, 2021

The EuroHPC Joint Undertaking (JU) was formalized in 2018, beginning a new era of European supercomputing that began to bear fruit this year with the launch of several of the first EuroHPC systems. The undertaking, however, has not been without its speed bumps, and the Union faces an uphill... Read more…

How Argonne Is Preparing for Exascale in 2022

September 8, 2021

Additional details came to light on Argonne National Laboratory’s preparation for the 2022 Aurora exascale-class supercomputer, during the HPC User Forum, held virtually this week on account of pandemic. Exascale Computing Project director Doug Kothe reviewed some of the 'early exascale hardware' at Argonne, Oak Ridge and NERSC (Perlmutter), while Ti Leggett, Deputy Project Director & Deputy Director... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer called Dojo to process truly vast amounts of video data. It’s a beast! … A truly useful exaflop at de facto FP32.” Read more…

Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

May 27, 2021

A ribbon-cutting ceremony held virtually at Berkeley Lab's National Energy Research Scientific Computing Center (NERSC) today marked the official launch of Perlmutter – aka NERSC-9 – the GPU-accelerated supercomputer built by HPE in partnership with Nvidia and AMD. Read more…

Google Launches TPU v4 AI Chips

May 20, 2021

Google CEO Sundar Pichai spoke for only one minute and 42 seconds about the company’s latest TPU v4 Tensor Processing Units during his keynote at the Google I Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. “We’ve been scaling our neural network training compute dramatically over the last few years,” said Milan Kovac, Tesla’s director of autopilot engineering. Read more…

CentOS Replacement Rocky Linux Is Now in GA and Under Independent Control

June 21, 2021

The Rocky Enterprise Software Foundation (RESF) is announcing the general availability of Rocky Linux, release 8.4, designed as a drop-in replacement for the soon-to-be discontinued CentOS. The GA release is launching six-and-a-half months after Red Hat deprecated its support for the widely popular, free CentOS server operating system. The Rocky Linux development effort... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Iran Gains HPC Capabilities with Launch of ‘Simorgh’ Supercomputer

May 18, 2021

Iran is said to be developing domestic supercomputing technology to advance the processing of scientific, economic, political and military data, and to strengthen the nation’s position in the age of AI and big data. On Sunday, Iran unveiled the Simorgh supercomputer, which will deliver.... Read more…

Leading Solution Providers

Contributors

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

Intel Launches 10nm ‘Ice Lake’ Datacenter CPU with Up to 40 Cores

April 6, 2021

The wait is over. Today Intel officially launched its 10nm datacenter CPU, the third-generation Intel Xeon Scalable processor, codenamed Ice Lake. With up to 40 Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire