Architecting for AI Workloads

March 4, 2019

Artificial intelligence has come of age. To capitalize fully on the opportunities, organizations need to design high-performance computing architectures for AI workloads.

After years of talking about the promise of artificial intelligence, enterprises around the world are now diving headfirst into AI-driven processes and business models. From financial services to manufacturing, from healthcare to retail, enterprises are now “all in” with AI and its supporting computing models — notably machine and deep learning. The same holds true for universities and government agencies. They are using AI for countless pursuits, from driving groundbreaking scientific discoveries to protecting our national security.

This widespread embrace of all things AI is fueled by the rise of more powerful processors and accelerators, advanced tools and techniques for data analytics, more precise algorithms and — most of all — an explosion of data, driven to a large degree by the Internet of Things. When you put it all together, you’ve got what it takes to put AI to work in countless applications.

Architecting for AI workloads

To capitalize fully on the opportunities in today’s data-driven world, IT organizations need to design high-performance computing architectures to accommodate demanding AI workloads. The HPC and AI community has started optimizing AI frameworks and developer tools to address performance needs, allowing for much larger batch sizes to be processed on industry standard CPUs. Within the last year, Intel® has seen up to 241x training performance gains through optimized frameworks with Intel® Math Kernel Library (MKL) on Intel Xeon® Scalable Processors over Haswell processors. This can take your time to train from hours to minutes, while these optimizations provide the eco-system greater access to AI capabilities.

This shift to AI-focused infrastructure is happening today as organizations roll out systems that bring together the capabilities of HPC, data analytics and AI. This is the case with the University of Cambridge’s latest supercomputer, called Cumulus. This groundbreaking system was designed to serve as a single HPC cluster that supports researchers’ needs for data analytics, machine learning and large-scale data processing. The goal is to solve extremely difficult big data, simulation and AI challenges.

To meet this goal, the Cumulus architecture was designed to address the broad range of system challenges, including those at the compute, network, storage and software layers. A key objective was to make the infrastructure perform well for diverse, data-intensive research workloads.

The Cumulus system provides more than 2 petaflops of performance, powered by Dell EMC PowerEdge™ servers and Intel Xeon Scalable processors, all connected via the Intel Omni-Path Architecture (OPA). The system incorporates OpenStack® software to control pools of compute, storage and networking resources and make them readily accessible to users via a cloud interface.

Solving for I/O bottlenecks

This architectural foundation alone doesn’t necessarily solve today’s persistent I/O challenges in HPC clusters. Here’s the problem: While data-processing power has raced forward in recent years, storage I/O limitations have created bottlenecks that slow time to insight, particularly for researchers running data-centric workloads that interact continuously with data storage systems.

The Cumulus system removes these bottlenecks with a unique solution called the Data Accelerator (aka DAC), which is designed into the network topology. DAC incorporates technologies from Dell EMC, Intel  and Cambridge University. In this architecture, the DAC nodes work in conjunction with the Distributed Name Space (DNE) feature in the Lustre file system and Intel® Omni-Path switches to accelerate system I/O.

The results of this accelerated architecture have been rather amazing. With DAC under the hood, Cumulus provides more than 500 GB/s of I/O read performance, which makes it the UK’s fastest HPC I/O platform, according to the university’s Research Computing Service, which operates the Cumulus cluster.[1]

In benchmark testing, the Cumulus system achieved an IO-500 score of 158.7, which ranked the system third on the November 2018 IO-500 list. For system users, these numbers equate to big improvements in I/O performance for data-intensive HPC and AI workloads — and faster time to insight.

Building the right foundation for new and emerging workloads

For organizations searching for the right IT foundation for AI workloads, Intel offers expert insights in its high-level Guide to Developing an AI Infrastructure Strategy. The options outlined in this guide range from starting from scratch with your current systems to outsourcing your entire solution. One of these options is to build a broad platform that is designed to support a wide range of AI workloads — which is the approach the University of Cambridge took with its Cumulus system.

The guide explains: “This approach is similar to the emerging ‘platform’ architecture we now see prevalent across IT — that is, an approach that provides a highly scalable infrastructure layer that can be managed as a single pool, using virtualization and software-defined orchestration across server processing, storage and networking.”[2]

The guide presents this broad-platform infrastructure strategy in terms of a three-tier stack, with hardware, software and process layers that work together to enable AI workloads. A few highlights from this architecture:

  • At the hardware layer, communication between devices and systems is based around an ultra-high speed backbone, such as the Intel® Omni-Path.
  • The software layer includes operating system and virtualization layers, which support a library of AI-specific modules. These modules enable algorithmic processing and analytics, data management and I/O, as well as the delivery of data sources and the visualization of analysis results.
  • The process layer runs the business logic of the AI application, using library modules to deliver capabilities like image recognition.

Intel notes that this architecture results in a platform-based approach that offers a single point of configuration and a unique deployment target.

Key takeaways

The rise of artificial intelligence creates unprecedented opportunities for today’s enterprises. To fully capitalize on these opportunities, your organization needs a scalable HPC infrastructure that is specifically designed to incorporate the latest processor and fabric technologies, accommodate massive amounts of data, and leverage technologies to accelerate the data storage I/O and AI workloads.

To learn more

For a closer and more technical look at the University of Cambridge’s use of the Data Accelerator, visit the Research Computing Services’ Data Accelerator site. And for a broader look at the university’s Cumulus cluster, read the Dell EMC case study “UK Science Cloud.”

 

The Convergence of HPC, Analytics and AI

High-performance computing, data analytics and artificial intelligence no longer live in separate domains. These complementary technologies are rapidly converging as organizations work to gain greater value from the data they capture and store.


[1] Dell EMC case study, “UK Science Cloud,” November 2018.

 

[2] Intel, “Select the Best Infrastructure Strategy to Support Your AI Solution,” March 2018.

 

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