Enabling Research with MATLAB on the TeraGrid

By Nicole Hemsoth

October 18, 2010

Rajesh Bhaskaran at Cornell’s Space Systems Design Studio CUSat Satellite Project is leading a multi-year effort to create and deploy an autonomous in-orbit inspection satellite system using a MATLAB-based simulation.

Meanwhile, Ricky Harjanto at UC San Diego’s Cartilage Tissue Engineering Lab is also using MATLAB to examine changes in the shape of mice femurs during postnatal development via statistical shape modeling techniques to determine variations in mouse development at different stages of growth.

At the same time, Harshal Mahajan at the University of Pittsburgh’s NSF Quality of Life Technology Center is modeling power wheelchair driving to determine different techniques to improve and enhance mobility for the many thousands who rely on safe, effective wheelchairs. Mahajan’s code uses the MATLAB system identification toolbox to build models from the wealth of driving data collected.

Outside of using MATLAB as their primary tool, these and other researchers have something else in common; they are all using Cornell University’s MATLAB on the TeraGrid experimental computing resource, which is helping them achieve fast results delivered to their desktop — and doing so in an operating environment they are already comfortable with.

High-Level Programming for the Non-Programmer

MATLAB is ubiquitous in scientific and large-scale computing with estimates closing in on over one million researchers who use the tool for a wide variety of technical computing applications. Outside of its use in technical applications, it is also being deployed to manipulate data gathered from a range of scientific instruments, including satellites, telescopes and sensors.

There are clear incentives to deliver easily-accessible software and computational resources to a large number of scientific users in general. This has been the goal of any number of universities and national labs from the era of grid until the present. This has been an aim of the National Science Foundation, which is one of a handful of funding sources for these types of projects and accordingly, it is not difficult to see how their interest was engaged when Cornell stated it would be capable of delivering MATLAB and high-performance computing to more researchers.

As Robert Buhrman, Senior Vice Provost for Research at Cornell, stated, “MATLAB on the TeraGrid will help enable a broader class of researchers who are well-versed in MATLAB to reduce the time to solution in a scalable manner without having to become parallel programming experts.” It is this reduced time to results and mitigation of programming challenges that makes this an attractive option — and one that has some direct results, judging from Cornell’s long list of research projects both pending and underway on the MATLAB and TeraGrid resource.

Part of the appeal for researchers is that the computational learning curve is diminished. Access to the 512-core resource does not require understanding of any particular operating system, MPI library, or batch scheduler. By utilizing the Parallel Computing Toolbox and the MATLAB Distributed Computing Server to access the resource via desktops and the TeraGrid science gateways, users who are part of TeraGrid are granted high-performance equipment without some of the common hassles on the programming front they used to encounter on a regular basis. In other words, it is allowing researchers to focus distinctly on their research problems, rather than forcing them to become, by proxy, experts in parallel programming.

The Partnership to Bring MATLAB to TeraGrid

Cornell University, in partnership with Purdue University, received an NSF grant to deploy MATLAB on the TeraGrid for what is currently deemed an experimental resource. Since MATLAB is such an important data tool for complex data analysis for many TeraGrid users, as a parallel resource it could provide an even greater opportunity to expand access to high-performance computing for researchers.

The goal of the partnership between the universities and the NSF is to provide “seamless parallel MATLAB computational services running on Windows HPC Server 2008 to remote desktop and Science Gateway users with complex analytic and fast simulation requirements.”

In a recent interview, David Lifka, director at the Cornell Center for Advanced Computing, noted that the funding from the NSF was in part to provide staff at Cornell that would develop software to allow MATLAB clients from any platform (Windows, Linux, Mac) to seamlessly connect to the experimental resource at Cornell and run jobs in parallel. This would mean that users would get their results back on their desktop via the Web interface without needing to learn a new batch system or new programming model. As Lifka explained, “Basically, once the users know MATLAB, they can use parallel MATLAB directly from their host client.”

The NSF also set aside funding for staff at Purdue University who were tasked with enabling the same sort of connectivity via their science gateway. Purdue has a software framework for building scientific gateways called HubZero — a framework that has been rising in popularity as more disciplines create domain-specific gateways of their own to share and augment research projects.

On a hardware and software level, it should be noted that Cornell’s cluster is not a “tricked out” resource by any means. The Dell PowerEdge HPC cluster is not a gigantic system; there are no special interconnects and it is not running any specialized, customized software. One look at the specs reveals that it’s running everything off the shelf, including Microsoft Windows HPC scheduler and the standard version of the MathWorks software, for example.

Lifka stated that the only part that is customized is the software interface that the client installs on his or her MATLAB client that handles the secure communication with the cluster to submit jobs.

The resource itself is modest, although the team hopes that it will eventually grow after proven success with the MATLAB on TeraGrid project. Current wait times are still an issue; this is not the instant-run access that some HPC-as-a-service providers from the “outside world” can deliver. The team publishes the current wait times, which generally run between three and four days, give or take.

Opening Doors to Discovery

MATLAB is in such wide use across disciplines because it allows researchers to focus on their immediate discipline-specific questions without needing to become advanced programmers. It is generally perceived as being far more compact for scientific and mathematical uses than Fortran or C, and for this reason, it is has become the most comfortable environment for many in academia, engineering and beyond. By delivering it to a larger number of users, Cornell, Purdue and TeraGrid are helping to advance scientific discovery and aid in the ease of access to many researchers.

“One of the beauties of MATLAB is that it’s such a broad tool that can be used across disciplines and that was the key thing we felt was important — and why we wanted to do this project with the NSF,” said Lifka. “The MathWorks’ MATLAB is used across business, academia and in national labs because it works and because it doesn’t require a steep learning curve. If you know your science and you know your MATLAB, you can get a lot done very quickly.”

Encouraging Broader Impact

Delivering parallel MATLAB as a resource for a broader class of researchers was part of what made the deal attractive to the National Science Foundation (NSF) as it examined the benefits of funding such a partnership. David Lifka, director at the Cornell Center for Advanced Computing, stated, “What we wanted to do and what the NSF wants to encourage is broader impact — bringing new users into the fold who need large-scale computing without the learning curve. We want to get them scaling their science up and hopefully, along the way, they ask some questions so we can continue to improve.”

The funding came from a Strategic Technologies and Cyberinfrastructure grant, which is backed by the NSF’s stated aims to bring new resources to bear to encourage greater access to high-performance computing. The idea behind the project is to present this as a resource so that later it can be determined whether or not this project will belong in the TeraGrid resource provider collection in the future. As Lifka noted, “We’re hopeful that someday we will be part of this collection, but today we’re not.”

Additional support for the project came from Dell, Microsoft and The Mathworks, purveyors of MATLAB. According to Lifka, this backing was due to the interest these stakeholders had in watching how utility computing could be made available and how the experimental resource might enable seamless access from Web to the desktop.

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