AI Systems Summit Keynote: Brace for System Level Heterogeneity Says de Supinski

By John Russell

April 1, 2021

Heterogeneous computing has quickly come to mean packing a couple of CPUs and one-or-many accelerators, mostly GPUs, onto the same node. Today, a one-such-node system has become the standard AI server offered by dozens of vendors. This is not to diminish the many advances (interconnect, memory, software, packaging, etc.) critical to delivering these powerful systems. Assemble a great many heterogeneous nodes together and, voila, you have a modern supercomputer. Think Summit and a host of other recent systems.

Now, the rise of new AI chips is prompting a re-think of that formula, at least in large-scale environments. Lawrence Livermore National Laboratory (LLNL) is a good example of early efforts to explore what’s being called System Level Heterogeneity. In his opening keynote at the AI Systems Summit this week, Bronis de Supinski, CTO of LLNL’s Livermore Computing Center, described how the lab has integrated two AI-specific systems – one from Cerebras and one from SambaNova – into two existing LLNL systems (Lassen (~23 petaflops) and Corona (~10 petaflops)) to achieve system level heterogeneity.

Bronis de Supinski, LLNL

The early results are impressive said de Supinski, citing a ~5x performance improvement per transistor over GPUs[i] for both the Cerebras and SambaNova solutions. For one application run on Cerebras he reported a 37x improvement per Volta GPU and 15x improvement per compute node. Importantly, he said LLNL is just getting started in its system level heterogeneity exploration and added the forthcoming El Capitan exascale system being built at LLNL will have elements of system level heterogeneity though he did not go into great detail.

Leaving aside the richness of resources the national labs possess, de Supinski came across as a believer in the new direction and set stage nicely in his early remarks:

“[W]e build a lot of supercomputers at Livermore Computing and many of those systems are heterogeneous, but they’re only heterogeneous at the node level. However, we are seeing a large number of very interesting devices being developed today, specifically things to accelerate AI portions of our workload. Frankly, it would not be cost effective for us to deploy these resources on every compute node,” he said. “Partly, that’s because our workloads are primarily physics-based simulations and many of our jobs would make no use of those resources. They would sit idle much of the time. In addition, many of these accelerators are, frankly, not designed to be deployed one per compute node; they are more expensive than we want to put on every compute node [and] more importantly, they’re would not be most effective [deployed] that way.

“[This has] led us to evolving our system design for these new workloads that we see, in particular one that we call cognitive simulation [which] we’re focused on today. We’re also looking at ways to develop [capabilities] for in situ or in transit data analysis. What this ends up leading us to is something that we call system level heterogeneity, which involves a system with diverse node types or even diverse resource types by design,” said de Supinski.

De Supinski didn’t downplay the challenges.

“Systems level heterogeneity raises many questions that we haven’t dealt with in our previous design system architecture designs. So how do we partition the work and handle the interactions across these diverse types of resources? What is the right mix of these different resource types? A particularly important question is as we disaggregate these resources, will the network bandwidth be sufficient and the network latency be sufficient for us to bring them together to serve the overall needs of individual jobs? Right?

“Can we have a job that uses these resources that are spread across the network? And this leads us to looking at how do we need to adapt our applications to use heterogeneity at the system level. Likely, this requires significantly more asynchrony, the ability to have one aspect of a simulation going on, concurrently to another, and then having them come together at perhaps even individual time step basis to then mix the results of those different pieces of that of the application. We are actively pursuing these directions at Livermore,” he said.

Much of the Cerebras material presented was familiar, reflecting de Supinski’s fall presentation at the AI Hardware Summit. (See HPCwire coverage, LLNL, ANL and GSK Provide Early Glimpse into Cerebras AI System Performance). Both the CS1-Lassen and SN10-Corona ‘systems’ use an abundance of InfiniBand pipes to connect to each other. The CS-1 system is hot – perhaps not a surprise given the giant size of its wafer-scale ‘chip’. “It’s got 400,000 AI optimized cores. It’s actually a 20-kilowatt device. So in fact, it’s one of the hottest compute devices on the planet,” said de Supinski.

On the issue of whether even with advanced connectivity at all levels the systems can talk to each other quickly enough, de Supinski was encouraged thus far but these are early days.

Both projects relied heavily on help from “What we call an Artificial Intelligence Center of Excellence (AI CoE). This is where we closely work with experts at Cerebras or SambaNova so that our applications can make use of these unique resources as part of our simulation workloads,” said de Supinski.

He said LLNL hasn’t had sufficient time to finely characterize the differences in strengths and weaknesses between the CS-1 and SN10-8R. “We know that they have some different capabilities in terms of the hardware and software. Some of the hardware we’re seeing is a little more maybe advanced with the Cerebras stuff, some of the software aspects or maybe further along with SambaNova, but they’re fairly similar,” said de Supinski; he estimates by SC21 they should have clearer sense of how the systems compare.

De Supinski is impressed with SambaNova’s multi-tenancy capabilities.

“An interesting aspect of the SN10-8R is that it’s got good support for doing multi tenancy, which is an aspect in the long-term we absolutely have to have support for in order to use this system level heterogeneity today. So now, we can run different jobs on that same resource, and just as importantly, we can run different models or different copies of the same model, and be able to actually send inference requests from different compute nodes to those different models and have it quickly process them and send them back,” he said.

De Supinski dug a little deeper into how the two systems handle a inertial confinement fusion simulation run at LLNL. A persistent challenge is adequately simulating at the needed time-steps and being able to modify model parameters. For this exercise, the AI portion of the model calculation is done on the specialized AI resource.

“On the CS-1, we’re doing approximately 18 million samples per second. (See slide below) Those samples take input from multiple zones running on individual compute nodes, [and] each of those zones involves 42 single-precision numbers, requiring about 3.2 gigabytes per second per zone. We’re then able to get those answers to come back, and those result in three interpolated scalars per zone. So less data, but still a fairly significant amount of data per time-step. Running over the overall simulation, what we see is we get roughly 37x higher performance than we get running on a single Volta and more than 15 times higher performance [that] running on individual compute nodes – a fairly significant performance boost.

Running a similar job on the SN10-8R, he said, “Here (slide below) we’re looking at the effect of batch size. So when you look at what we’re doing, we’re actually sending fairly small batches. And a key aspect of these AI accelerators is that they provide significantly higher performance on small batch sizes. So we’re, we’re really looking at running in this kind of region of the batch sizes (steep curve section) and we can get significant throughput, even on these small numbers of samples. Since we’re able to support multiple models on the SN10-8r, we can actually go about getting enough performance to justify sending the data across the network, running the AI model, and getting the neural network getting the results of that neural network back in time to continue with that within each time-step.”

Looking ahead, de Supinski offered a few thoughts on both El Capitan and LLNL’s forthcoming CTS-2 system procurement.

“El Capitan will be a little bit later than then the Frontier system. There’ll be a slightly different architecture as well. One of the differences I’ve spoken about recently is that we’ll be deploying something we call Rabbit modules, which allow us to essentially deploy data nodes that can work to speed up file system performance, but also to allow us to do some data analysis before we actually push things all the way back to our parallel file system,” said de Supinski. (see HPCwire coverage, Livermore’s El Capitan Supercomputer to Debut HPE ‘Rabbit’ Near Node Local Storage)

“We are also in the process of running our CTS-2 RFPs – the ASC CTS 2 procurement. The responses have been back for a while now, and expect the announcement of what exactly we’re getting under CTS-2 in general to come in about two months from now. We expect under that contract to continue exploring system level heterogeneity. We will be deploying similar systems under that [which] allow us to look at cognitive simulation and the benefits of system level heterogeneity as well as other systems that will support vastly improved data analysis as part of the overall workload,” he said.

“In order to support the system level heterogeneity, you need some advanced advances in the system software. A very important aspect of that is that we’re developing the next generation resource manager that we call Flux. So rather than running Slurm or [what we currently] run on, Lassen and Sierra that I won’t mentioned by name, we will be running Flux on the CTS-2 systems and on El Capitan. This resource manager allows us to treat different resources in different ways and to easily allocate these diverse resources and recognizing locality between them.”

De Supinski’s keynote was quick-moving but substantive and worth watching. Link to video: https://vimeo.com/531218771/b7d9b279a2

Editor’s Note:
AI Systems Summit organizer, Kiasco Research, has made the video of de Supinski’s keynote freely available. Free on demand passes to recordings of the other sessions “are available to members of research institutions, government agencies, and national labs, whilst corporate entities can purchase a pass for a small fee.”

[i] de Supinski said the performance improvement per transistor metric is preferred because it bakes in several items and provides a better measure of “performance per dollar we’re achieving.”

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!

ISC Keynote: Glimpse into Microsoft’s View of the Quantum Computing Landscape

June 15, 2021

Looking for a dose of reality and realistic optimism about quantum computing? Matthias Troyer, Microsoft distinguished scientist, plans to do just that in his ISC2021 keynote in two weeks –  Quantum Computing: From Ac Read more…

Q&A with Google’s Bill Magro, an HPCwire Person to Watch in 2021

June 11, 2021

Last Fall Bill Magro joined Google as CTO of HPC, a newly created position, after two decades at Intel, where he was responsible for the company's HPC strategy. This interview was conducted by email at the beginning of A Read more…

A Carbon Crisis Looms Over Supercomputing. How Do We Stop It?

June 11, 2021

Supercomputing is extraordinarily power-hungry, with many of the top systems measuring their peak demand in the megawatts due to powerful processors and their correspondingly powerful cooling systems. As a result, these Read more…

Honeywell Quantum and Cambridge Quantum Plan to Merge; More to Follow?

June 10, 2021

Earlier this week, Honeywell announced plans to merge its quantum computing business, Honeywell Quantum Solutions (HQS), which focuses on trapped ion hardware, with the U.K.-based Cambridge Quantum Computing (CQC), which Read more…

ISC21 Keynoter Xiaoxiang Zhu to Deliver a Bird’s-Eye View of a Changing World

June 10, 2021

ISC High Performance 2021 – once again virtual due to the ongoing pandemic – is swiftly approaching. In contrast to last year’s conference, which canceled its in-person component with a couple months’ notice, ISC Read more…

AWS Solution Channel

Building highly-available HPC infrastructure on AWS

Reminder: You can learn a lot from AWS HPC engineers by subscribing to the HPC Tech Short YouTube channel, and following the AWS HPC Blog channel. Read more…

Xilinx Expands Versal Chip Family With 7 New Versal AI Edge Chips

June 10, 2021

FPGA chip vendor Xilinx has been busy over the last several years cranking out its Versal AI Core, Versal Premium and Versal Prime chip families to fill customer compute needs in the cloud, datacenters, networks and more. Now Xilinx is expanding its reach to the booming edge... Read more…

ISC Keynote: Glimpse into Microsoft’s View of the Quantum Computing Landscape

June 15, 2021

Looking for a dose of reality and realistic optimism about quantum computing? Matthias Troyer, Microsoft distinguished scientist, plans to do just that in his I Read more…

A Carbon Crisis Looms Over Supercomputing. How Do We Stop It?

June 11, 2021

Supercomputing is extraordinarily power-hungry, with many of the top systems measuring their peak demand in the megawatts due to powerful processors and their c Read more…

Honeywell Quantum and Cambridge Quantum Plan to Merge; More to Follow?

June 10, 2021

Earlier this week, Honeywell announced plans to merge its quantum computing business, Honeywell Quantum Solutions (HQS), which focuses on trapped ion hardware, Read more…

ISC21 Keynoter Xiaoxiang Zhu to Deliver a Bird’s-Eye View of a Changing World

June 10, 2021

ISC High Performance 2021 – once again virtual due to the ongoing pandemic – is swiftly approaching. In contrast to last year’s conference, which canceled Read more…

Xilinx Expands Versal Chip Family With 7 New Versal AI Edge Chips

June 10, 2021

FPGA chip vendor Xilinx has been busy over the last several years cranking out its Versal AI Core, Versal Premium and Versal Prime chip families to fill customer compute needs in the cloud, datacenters, networks and more. Now Xilinx is expanding its reach to the booming edge... Read more…

What is Thermodynamic Computing and Could It Become Important?

June 3, 2021

What, exactly, is thermodynamic computing? (Yes, we know everything obeys thermodynamic laws.) A trio of researchers from Microsoft, UC San Diego, and Georgia Tech have written an interesting viewpoint in the June issue... Read more…

AMD Introduces 3D Chiplets, Demos Vertical Cache on Zen 3 CPUs

June 2, 2021

At Computex 2021, held virtually this week, AMD showcased a new 3D chiplet architecture that will be used for future high-performance computing products set to Read more…

Nvidia Expands Its Certified Server Models, Unveils DGX SuperPod Subscriptions

June 2, 2021

Nvidia is busy this week at the virtual Computex 2021 Taipei technology show, announcing an expansion of its nascent Nvidia-certified server program, a range of Read more…

AMD Chipmaker TSMC to Use AMD Chips for Chipmaking

May 8, 2021

TSMC has tapped AMD to support its major manufacturing and R&D workloads. AMD will provide its Epyc Rome 7702P CPUs – with 64 cores operating at a base cl 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…

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…

CERN Is Betting Big on Exascale

April 1, 2021

The European Organization for Nuclear Research (CERN) involves 23 countries, 15,000 researchers, billions of dollars a year, and the biggest machine in the worl 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…

HPE Launches Storage Line Loaded with IBM’s Spectrum Scale File System

April 6, 2021

HPE today launched a new family of storage solutions bundled with IBM’s Spectrum Scale Erasure Code Edition parallel file system (description below) and featu Read more…

Quantum Computer Start-up IonQ Plans IPO via SPAC

March 8, 2021

IonQ, a Maryland-based quantum computing start-up working with ion trap technology, plans to go public via a Special Purpose Acquisition Company (SPAC) merger a Read more…

Leading Solution Providers

Contributors

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…

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…

AMD Launches Epyc ‘Milan’ with 19 SKUs for HPC, Enterprise and Hyperscale

March 15, 2021

At a virtual launch event held today (Monday), AMD revealed its third-generation Epyc “Milan” CPU lineup: a set of 19 SKUs -- including the flagship 64-core, 280-watt 7763 part --  aimed at HPC, enterprise and cloud workloads. Notably, the third-gen Epyc Milan chips achieve 19 percent... Read more…

Can Deep Learning Replace Numerical Weather Prediction?

March 3, 2021

Numerical weather prediction (NWP) is a mainstay of supercomputing. Some of the first applications of the first supercomputers dealt with climate modeling, and Read more…

Livermore’s El Capitan Supercomputer to Debut HPE ‘Rabbit’ Near Node Local Storage

February 18, 2021

A near node local storage innovation called Rabbit factored heavily into Lawrence Livermore National Laboratory’s decision to select Cray’s proposal for its CORAL-2 machine, the lab’s first exascale-class supercomputer, El Capitan. Details of this new storage technology were revealed... Read more…

GTC21: Nvidia Launches cuQuantum; Dips a Toe in Quantum Computing

April 13, 2021

Yesterday Nvidia officially dipped a toe into quantum computing with the launch of cuQuantum SDK, a development platform for simulating quantum circuits on GPU-accelerated systems. As Nvidia CEO Jensen Huang emphasized in his keynote, Nvidia doesn’t plan to build... Read more…

Microsoft to Provide World’s Most Powerful Weather & Climate Supercomputer for UK’s Met Office

April 22, 2021

More than 14 months ago, the UK government announced plans to invest £1.2 billion ($1.56 billion) into weather and climate supercomputing, including procuremen Read more…

African Supercomputing Center Inaugurates ‘Toubkal,’ Most Powerful Supercomputer on the Continent

February 25, 2021

Historically, Africa hasn’t exactly been synonymous with supercomputing. There are only a handful of supercomputers on the continent, with few ranking on the Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire