Weather at Exascale: Load Balancing for Heterogeneous Systems

By Oliver Peckham

March 30, 2020

The first months of 2020 were dominated by weather and climate supercomputing news, with major announcements coming from the UK, the European Centre for Medium-Range Weather Forecasts and the U.S. National Oceanic and Atmospheric Administration. Richard Loft, director of technology development at the Computational Information Systems Laboratory of the NSF-supported National Center for Atmospheric Research (NCAR), took the (virtual) stage at Nvidia’s GTC 2020 to highlight NCAR’s progress in moving weather and climate supercomputing into the exascale era.

Specifically, Loft discussed NCAR’s Model for Prediction Across Scales (MPAS), its latest-generation weather model. MPAS, developed in partnership with Los Alamos National Laboratory, is the basis for the Global High-Resolution Forecasting System (GRAF) introduced last year by IBM and its subsidiary The Weather Company and running in production since October. Loft, who noted that the team was funding-limited and relied heavily on student support, called GRAF the “crowning achievement” of MPAS so far. 

“This partnership between NCAR, IBM and the Weather Company,” Loft said, “has put into production a model with three-kilometer resolution over lots of areas in the world – and especially the developing world – where they haven’t had those kinds of resolutions before.” These new capabilities, he said, would enable significantly enhanced weather forecast products for those underserved areas.

And, crucially: “[GRAF] is the first GPU-based global forecast model, to my knowledge, in production anywhere in the world at this point.”

The goals for MPAS

In his talk, Loft outlined the objectives NCAR is aiming for with MPAS as the computing world approaches the exascale era – and how load balancing between GPUs and CPUs factors into those objectives, with GPUs becoming an increasingly important part of many major supercomputers. “So when we started this over three years ago,” Loft said, “we started looking at it from the perspective of trying to achieve a number of core capabilities for this model.”

First was performance portability. “We wanted to get the best performance on GPUs we could, but we wanted to maintain performance on CPUs,” Loft said. “We didn’t want to sacrifice that, because that’s our bread and butter.” 

Second was resilience. “We needed to have a strategy for load-balancing all the code on the hybrid node,” he explained, “in such a way that it wasn’t too hard-wired for any one combination of CPU and GPU.”

Finally, the researchers wanted to port the minimum amount of code. For this, Loft said, they “triaged” the process. They ported most of the physics and dynamics code to the GPUs (using OpenACC), but left the radiation code (which accounts for the transfer of solar radiation through the atmosphere) as a CPU-only code. They made this choice, Loft explained, because the radiative code was over 30,000 lines. Radiation, Loft said, was “expensive”: “If you called it every timestep, the model would grind to a stop.” Fortunately, radiation also evolved on a much slower timescale, so operating it asynchronously from the rest of the model was sensible on a sheer physical level.

The researchers also left the included land surface model – which accounts for water and energy transfers between land and the atmosphere – as a CPU-only model, since it was over 20,000 lines of “branchy code” and didn’t take much time to run (unfortunately, though, Loft said, it had to be run synchronously).

Tinkering to perfection on two powerful supercomputers

The Summit supercomputer.

To optimize its newly ported model, NCAR turned to a pair of supercomputers. First was NCAR’s own Cheyenne system, which houses 4,032 nodes (each with Intel Xeon Broadwell processors), an aggregate 313 TB of memory, connected by Mellanox EDR InfiniBand. Second was Summit at Oak Ridge National Laboratory, housing 4,608 nodes (each with two IBM Power9 CPUs and six Nvidia Volta GPUs), over 10 PB of memory, tied together with EDR InfiniBand. 

At 4.8 and 148.6 Linpack petaflops, Cheyenne and Summit respectively placed 44th and 1st on the most recent Top500 list of the world’s most powerful publicly ranked supercomputers. The two systems were carefully selected: Cheyenne was selected for its CPU-only homogeneous design, while Summit offered a glimpse into the future of heterogeneous, GPU-heavy system design. Summit also bore similarity to GRAF’s IBM Power9-based supercomputer, which features 84 nodes with four Nvidia V100 GPUs each.

The supercomputers were employed to help NCAR find the “Goldilocks zone” for resource allocation. When allocating CPU cores and/or GPUs to either the radiation or the dynamics or the radiation, a mismatch in the timing could lead to serious losses of time and money. Allocate too little to radiation on the CPUs, for instance, and the GPUs powering integration (the remainder of the model) would have to pause their work while waiting for the CPUs to catch up.

An illustration of the possible mismatches between model timing. Image courtesy of NCAR.

For the tests, the researchers leveraged 76 CPU nodes on Cheyenne and 76 hybrid nodes on Summit to run MPAS at a ten-kilometer resolution, throwing 81,920 points of analysis to each node on each system. They tested balances in CPU and GPU allocation, changes in symmetric multithreading and changes in the interval at which radiation was re-integrated into the model. 

On Cheyenne, the researchers found that the system was integration-limited, performing relatively poorly on general model tasks – so, they assigned most of the cores on each CPU to the general model and very few to radiation (a 2:1 ratio). On Summit, meanwhile, they found that with the GPUs contributing massive amounts to the processing power, the system was radiation-limited. Eventually, the researchers found the best performance at a balance of 18 CPU cores (and the GPUs) dedicated to the integrated model, with 24 dedicated to radiation.

Summit’s configuration for MPAS. Image courtesy of NCAR.

In addition to finding these settings for optimization, the researchers saw three times the throughput for MPAS when using GPU nodes compared to the CPU nodes. In terms of the frequency of re-integrating the radiation, Loft said that on Summit, they found that “there is insufficient CPU power to keep up with six GPUs unless the radiation [was] called pretty infrequently.”

The path forward

Next, the NCAR team will work on further optimizing message passing, as well as porting the land surface model to GPUs to free up even more cores for radiation processing. (Loft says that they’re considering porting the radiation code to GPUs, but only as a “last resort.) 

While Loft acknowledges that there’s work to be done on scalability and throughput, he says the model is well on its way to being optimized. “It’s a work in progress, but we’ve got the full model working,” Loft said. “It’s a good first cut.”

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!

HPC User Forum: Sustainability at TACC Points to Software

October 3, 2023

Recently, Dan Stanzione, Executive Director, TACC and Associate Vice President for Research, UT-Austin, gave a presentation on HPC sustainability at the Fall 2023 HPC Users Forum. The complete set of slides is available Read more…

Google’s Controversial AI Chip Paper Under Scrutiny Again 

October 3, 2023

A controversial research paper by Google that claimed the superiority of AI techniques in creating chips is under the microscope for the authenticity of its claims. Science publication Nature is investigating Google's c Read more…

Rust Busting: IBM and Boeing Battle Corrosion with Simulations on Quantum Computer

October 3, 2023

The steady research into developing real-world applications for quantum computing is piling up interesting use cases. Today, IBM reported on work with Boeing to simulate corrosion processes to improve composites used in Read more…

Nvidia Delivering New Options for MLPerf and HPC Performance

September 28, 2023

As HPCwire reported recently, the latest MLperf benchmarks are out. Not unsurprisingly, Nvidia was the leader across many categories. The HGX H100 GPU systems, which contain eight H100 GPUs, delivered the highest throughput on every MLPerf inference test in this round. Read more…

Hakeem Oluseyi Explores His Unlikely Journey from the Street to the Stars in SC23 Keynote

September 28, 2023

Defying the odds In the heart of one of the toughest neighborhoods in the country, young Hakeem Oluseyi’s world was a confined space, but his imagination soared to the stars. While other kids roamed the streets, he Read more…

AWS Solution Channel

Shutterstock 2338659951

VorTech Derisks Innovative Technology to Aid Global Water Sustainability Challenges Using Cloud-Native Simulations on AWS

Overview

No more than 1 percent of the world’s water is readily available fresh water, according to the US Geological Survey. Read more…

QCT Solution Channel

QCT and Intel Codeveloped QCT DevCloud Program to Jumpstart HPC and AI Development

Organizations and developers face a variety of issues in developing and testing HPC and AI applications. Challenges they face can range from simply having access to a wide variety of hardware, frameworks, and toolkits to time spent on installation, development, testing, and troubleshooting which can lead to increases in cost. Read more…

Nvidia Takes Another Shot at Trying to Get AI to Mobile Devices

September 28, 2023

Nvidia takes another shot at trying to get to mobile devices Long before the current situation of Nvidia's GPUs holding AI hostage, the company tried to put its chips in mobile devices but failed. The Tegra mobile chi Read more…

Shutterstock 1927423355

Google’s Controversial AI Chip Paper Under Scrutiny Again 

October 3, 2023

A controversial research paper by Google that claimed the superiority of AI techniques in creating chips is under the microscope for the authenticity of its cla Read more…

Rust Busting: IBM and Boeing Battle Corrosion with Simulations on Quantum Computer

October 3, 2023

The steady research into developing real-world applications for quantum computing is piling up interesting use cases. Today, IBM reported on work with Boeing to Read more…

Nvidia Delivering New Options for MLPerf and HPC Performance

September 28, 2023

As HPCwire reported recently, the latest MLperf benchmarks are out. Not unsurprisingly, Nvidia was the leader across many categories. The HGX H100 GPU systems, which contain eight H100 GPUs, delivered the highest throughput on every MLPerf inference test in this round. Read more…

IonQ Announces 2 New Quantum Systems; Suggests Quantum Advantage is Nearing

September 27, 2023

It’s been a busy week for IonQ, the quantum computing start-up focused on developing trapped-ion-based systems. At the Quantum World Congress today, the compa Read more…

Rethinking ‘Open’ for AI

September 27, 2023

What does “open” mean in the context of AI? Must we accept hidden layers? Do copyrights and patents still hold sway? And do consumers have the right to opt Read more…

Aurora Image

Leveraging Machine Learning in Dark Matter Research for the Aurora Exascale System 

September 25, 2023

Scientists have unlocked many secrets about particle interactions at atomic and subatomic levels. However, one mystery that has eluded researchers is dark matte Read more…

Watsonx Brings AI Visibility to Banking Systems

September 21, 2023

A new set of AI-based code conversion tools is available with IBM watsonx. Before introducing the new "watsonx," let's talk about the previous generation Watson Read more…

Intel’s Gelsinger Lays Out Vision and Map at Innovation 2023 Conference

September 20, 2023

Intel’s sprawling, optimistic vision for the future was on full display yesterday in CEO Pat Gelsinger’s opening keynote at the Intel Innovation 2023 confer Read more…

CORNELL I-WAY DEMONSTRATION PITS PARASITE AGAINST VICTIM

October 6, 1995

Ithaca, NY --Visitors to this year's Supercomputing '95 (SC'95) conference will witness a life-and-death struggle between parasite and victim, using virtual Read more…

SGI POWERS VIRTUAL OPERATING ROOM USED IN SURGEON TRAINING

October 6, 1995

Surgery simulations to date have largely been created through the development of dedicated applications requiring considerable programming and computer graphi Read more…

U.S. Will Relax Export Restrictions on Supercomputers

October 6, 1995

New York, NY -- U.S. President Bill Clinton has announced that he will definitely relax restrictions on exports of high-performance computers, giving a boost Read more…

Dutch HPC Center Will Have 20 GFlop, 76-Node SP2 Online by 1996

October 6, 1995

Amsterdam, the Netherlands -- SARA, (Stichting Academisch Rekencentrum Amsterdam), Academic Computing Services of Amsterdam recently announced that it has pur Read more…

Cray Delivers J916 Compact Supercomputer to Solvay Chemical

October 6, 1995

Eagan, Minn. -- Cray Research Inc. has delivered a Cray J916 low-cost compact supercomputer and Cray's UniChem client/server computational chemistry software Read more…

NEC Laboratory Reviews First Year of Cooperative Projects

October 6, 1995

Sankt Augustin, Germany -- NEC C&C (Computers and Communication) Research Laboratory at the GMD Technopark has wrapped up its first year of operation. Read more…

Sun and Sybase Say SQL Server 11 Benchmarks at 4544.60 tpmC

October 6, 1995

Mountain View, Calif. -- Sun Microsystems, Inc. and Sybase, Inc. recently announced the first benchmark results for SQL Server 11. The result represents a n Read more…

New Study Says Parallel Processing Market Will Reach $14B in 1999

October 6, 1995

Mountain View, Calif. -- A study by the Palo Alto Management Group (PAMG) indicates the market for parallel processing systems will increase at more than 4 Read more…

Leading Solution Providers

Contributors

CORNELL I-WAY DEMONSTRATION PITS PARASITE AGAINST VICTIM

October 6, 1995

Ithaca, NY --Visitors to this year's Supercomputing '95 (SC'95) conference will witness a life-and-death struggle between parasite and victim, using virtual Read more…

SGI POWERS VIRTUAL OPERATING ROOM USED IN SURGEON TRAINING

October 6, 1995

Surgery simulations to date have largely been created through the development of dedicated applications requiring considerable programming and computer graphi Read more…

U.S. Will Relax Export Restrictions on Supercomputers

October 6, 1995

New York, NY -- U.S. President Bill Clinton has announced that he will definitely relax restrictions on exports of high-performance computers, giving a boost Read more…

Dutch HPC Center Will Have 20 GFlop, 76-Node SP2 Online by 1996

October 6, 1995

Amsterdam, the Netherlands -- SARA, (Stichting Academisch Rekencentrum Amsterdam), Academic Computing Services of Amsterdam recently announced that it has pur Read more…

Cray Delivers J916 Compact Supercomputer to Solvay Chemical

October 6, 1995

Eagan, Minn. -- Cray Research Inc. has delivered a Cray J916 low-cost compact supercomputer and Cray's UniChem client/server computational chemistry software Read more…

NEC Laboratory Reviews First Year of Cooperative Projects

October 6, 1995

Sankt Augustin, Germany -- NEC C&C (Computers and Communication) Research Laboratory at the GMD Technopark has wrapped up its first year of operation. Read more…

Sun and Sybase Say SQL Server 11 Benchmarks at 4544.60 tpmC

October 6, 1995

Mountain View, Calif. -- Sun Microsystems, Inc. and Sybase, Inc. recently announced the first benchmark results for SQL Server 11. The result represents a n Read more…

New Study Says Parallel Processing Market Will Reach $14B in 1999

October 6, 1995

Mountain View, Calif. -- A study by the Palo Alto Management Group (PAMG) indicates the market for parallel processing systems will increase at more than 4 Read more…

ISC 2023 Booth Videos

Cornelis Networks @ ISC23
Dell Technologies @ ISC23
Intel @ ISC23
Lenovo @ ISC23
Microsoft @ ISC23
ISC23 Playlist
  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire