Building the Team: South African Style

By Dan Olds

June 9, 2019

We’re only eight days away from the start of the ISC 2019 Student Cluster Competition. Fourteen student teams from eleven countries will travel to Frankfurt, Germany, to vie for the ISC19 cluster competition championship trophy, highest Linpack award, and the Fan Favorite award. It’s going to be a very exciting event, chock full of triumph, with a smattering of tragedy.

We’ll be covering this competition in our usual comprehensive way from the opening bell all the way to the ending gong (note to self: bring a gong to Germany).

But we’re going to tell a different story today. It’s a story about how one organization is using the Student Cluster Competition to build a new work force, schooled in HPC, who will help to revolutionize the technical infrastructure for an entire country.

The Centre for High Performance Computing (CHPC), under the tutelage of Happy Sitole, has been participating in the ISC Student Cluster Competition for the past six years, winning the championship three times, taking second place twice, and winning the bronze medal once. That’s a fantastic record. What’s even more interesting is how they did it.

Cultivating Home Grown Talent

The overriding goal of the CHPC is to spread HPC knowledge far and wide throughout South Africa and, over time, build a workforce that understands HPC and can fuel a computing-led renaissance in the country. Participating in Student Cluster Competitions is one of the keys to making these goals a reality.

Most cluster competition teams start at the university level – a professor or a group of students hear about the competition, decide to put together a team and apply for admission to the tournament. South Africa takes a decidedly different approach.

The CHPC puts together basic HPC training sessions and offers them to any university in the country that’s interested. These classes start at a very basic level – the vast majority of the kids have never been exposed to HPC, Linux, or cluster computing concepts. Once they have initial training, each university decides if they want to field a team of four students and send them to the intra-country competition – traditionally held at the CHPC conference each December.

The team that wins the CHPC conference competition become the first four members of the CHPC national team. Judges then select two outstanding students from other teams, which brings the squad up to six strong.

100% Green

One of the important policies that the CHPC has instituted is that every CHPC team is composed of students who haven’t participated in a previous competition. Former team members mentor current teams and help prepare them for the competition in Frankfurt. While this constant turnover of student competitors might make it tougher for CHPC to compete with more experienced teams, it produces the maximum number of HPC-trained students who are ready for employment by industry or to pursue further education.

2018 Action & Results

Ten teams took up the challenge of the South African Student Cluster Competition, held in early December in Cape Town. The entire competition was generously underwritten by Dell, with the company providing a menu of equipment for the teams. This was also the first challenge for the kids; how should they configure their cluster?

They were given a modest budget and a price list, everything else was up to them. Most teams selected three beefy dual-processor workstations, with one acting as a head node and the others as compute nodes. They also tended to use a 10GbE point to point direct interconnect rather than spend money on a dedicated switch. This worked out for the most part, but nearly every team (except the two with switches) had interconnect problems at one point or another.

The teams faced a challenging set of benchmarks and scientific applications over the three-day event. On the benchmark side, they had to run HPCC (an amalgamation of several benchmarks), along with a separate HPC (Linpack) run, and the ever-grueling HPCG benchmark.

Team $Witts jumped out to an early lead with their pack-leading Linpack9210 score and their second place HPCC score. On HPCC, Team Department of Spaghetti Engineering topped the field, with Team $Witts, Team COMMITment Issues, Team Short Circuits, and Team Will Code for Pizza in a near photo finish for second and third places.

Department of Spaghetti Engineering also topped the rest of the teams on HPCG, narrowly edging out Short Circuits and $Witts.

Things got a lot harder when it came to the real-world scientific application runs. The first application was GROMACS and it contained four separate tasks. The first was a adh_cubic run, then 1.5m_water, a visualization, then doing a run on the Microsoft Azure Cloud configured with two NVIDIA V100 GPUs.

Team Will Code for Pizza grabbed first place on adh_cubic with a comfortable margin over Short Circuits, $Witts, and Department of Spaghetti Engineering. But the Short Circuits struck back on the 1.5m water run, pounding the rest of the field. Pretty much everyone successfully completed the visualization round, gaining maximum points for this task. The Short Circuits continued their winning streak by taking the top score on the Azure GPU GROMACS run, but Department of Spaghetti Engineering ran a very close second. Will Code for Pizza and $Witts were just an eyelash behind, taking home third and fourth place respectively.

At this point in the competition, Short Circuits had a slight lead over $Witts and Department of Spaghetti Engineering with Will Code for Pizza still in the hunt. But the toughest applications were still ahead.

FEniCS, an application used to solve partial differential equations, turned out to be the Waterloo for many teams. The way the CHPC cluster competition works is that teams are required to complete an application before getting the dataset for the next application. I think that many teams didn’t get their FEniCS dataset until it was too late for them to fully set up and optimize their run.

One team that didn’t have this problem was The Short Circuits. They nailed down yet another win on FEniCS, with Will Code for Pizza running in second with $Witts and Department of Spaghetti Engineering trailing the leaders. But there was still one application left to run:  Tensor Flow.

This turned out to be the “Application of Destiny” for the 2018 CHPC Student Cluster Competition. Only one team managed to finish this app run and nab all of the point for completing it. And that team was Department of Spaghetti Engineering. This gave them enough points to log a normalized overall score of 84.54%, taking home the championship trophy and setting their sights on the ISC19 competition in Frankfurt.

It was a close competition with The Short Circuits taking second with their score of 77.47%, $Witts in third with 72.58% and Will Code for Pizza getting honorable mention for their fourth-place finish of 69.36%.

Meet the Teams and See the Awards

We were in Cape Town with our plethora of video gear, including a new camera, to interview the teams and the awards ceremony. The team videos have some sound issues, which I’ll blame on myself for not adequately learning how to integrate the new camera into my gear stack. Still, take a look at the team videos and get to know them, they’ve got a lot of personality and will be great additions to our industry.

Department of Spaghetti Engineering:

The Short Circuits:

$Witts: 

Will Code for Pizza:

Tiger Team:  

Error 404:

Team Phoenix:

Team Thar:

COMMITment Issues:

Kalisa Academy:

Was There a Party? Oh Yes There Was!

If there’s only one thing I’ve learned by attending three CHPC conferences, it’s this:  the CHPC knows how to party. Combine fantastic food, plenty to drink, great entertainment, plus HPC student awards, and you have all the ingredients you need for one hell of a great time.

As part of the party, plenty of awards were handed out, along with generous gift bags for the winning teams and students. Here’s who won what:

  • Innovation Award, sponsored by Altair, for the best cluster design:  Team Error 404
  • Best Teamwork Award, sponsored by Bright Computing:  Team Will Code for Pizza
  • Beat the Clock Award, for the team that just barely got their results in, sponsored by Mellanox:  $Witts
  • Most Independent Team, the team that needed the least help from mentors, sponsored by Eclipse Holdings:  Team Short Circuits
  • Least Stressed Team, the team that had the most fun:  Department of Spaghetti Engineering
  • Best Female Competition Participant, a massive $5,000 award sponsored by Intel:  Mapule Madzena

Here’s a video of the awards section of the party:

 

In January, following the competition, Dell brought the entire team to Dell headquarters in Austin, Texas. While there, Dell engineers helped train the students and collaborated on the design of the team’s ISC19 system. The CHPC team were also guests at the Texas Advanced Computing Center, where they soaked up even more HPC knowledge.

This team should be well prepared for the ISC19 competition in Frankfurt next week. If they hold true to form, they’ll be on the podium when the awards are announced. One hallmark of the CHPC teams is that they tend to outwork their competitors at these events. We’ll see if that happens again this year, but they definitely be a force to be reckoned with in Frankfurt.

As always, we’ll be covering all of the action at ISC19, so stay tuned……

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!

Empowering High-Performance Computing for Artificial Intelligence

April 19, 2024

Artificial intelligence (AI) presents some of the most challenging demands in information technology, especially concerning computing power and data movement. As a result of these challenges, high-performance computing Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that have occurred about once a decade. With this in mind, the ISC Read more…

2024 Winter Classic: Texas Two Step

April 18, 2024

Texas Tech University. Their middle name is ‘tech’, so it’s no surprise that they’ve been fielding not one, but two teams in the last three Winter Classic cluster competitions. Their teams, dubbed Matador and Red Read more…

2024 Winter Classic: The Return of Team Fayetteville

April 18, 2024

Hailing from Fayetteville, NC, Fayetteville State University stayed under the radar in their first Winter Classic competition in 2022. Solid students for sure, but not a lot of HPC experience. All good. They didn’t Read more…

Software Specialist Horizon Quantum to Build First-of-a-Kind Hardware Testbed

April 18, 2024

Horizon Quantum Computing, a Singapore-based quantum software start-up, announced today it would build its own testbed of quantum computers, starting with use of Rigetti’s Novera 9-qubit QPU. The approach by a quantum Read more…

2024 Winter Classic: Meet Team Morehouse

April 17, 2024

Morehouse College? The university is well-known for their long list of illustrious graduates, the rigor of their academics, and the quality of the instruction. They were one of the first schools to sign up for the Winter Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that ha Read more…

Software Specialist Horizon Quantum to Build First-of-a-Kind Hardware Testbed

April 18, 2024

Horizon Quantum Computing, a Singapore-based quantum software start-up, announced today it would build its own testbed of quantum computers, starting with use o Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pre Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Instit Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips Read more…

The VC View: Quantonation’s Deep Dive into Funding Quantum Start-ups

April 11, 2024

Yesterday Quantonation — which promotes itself as a one-of-a-kind venture capital (VC) company specializing in quantum science and deep physics  — announce Read more…

Nvidia’s GTC Is the New Intel IDF

April 9, 2024

After many years, Nvidia's GPU Technology Conference (GTC) was back in person and has become the conference for those who care about semiconductors and AI. I Read more…

Google Announces Homegrown ARM-based CPUs 

April 9, 2024

Google sprang a surprise at the ongoing Google Next Cloud conference by introducing its own ARM-based CPU called Axion, which will be offered to customers in it 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…

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…

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…

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…

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…

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…

Leading Solution Providers

Contributors

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…

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…

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…

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…

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…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire