The Six Personalities of Supercomputing

By Andrew Jones

November 16, 2012

There are two types of people in supercomputing – people that have a top 10 supercomputer and people that don’t. Or people who understand the exascale problem and people who understand the missing middle problem. Or people who have scalable applications and people that don’t.

Or people who claim just two types of person and then list several non-exclusive options.

In my usual part serious, part provocative style, here is my light-hearted look at the different personality stereotypes involved in high performance computing. This is by no means an exclusive list, but it does illustrate the range of people who contribute to the flavor of the world of supercomputing.

1. The Great Wall of China type.Our supercomputer is so big you can see it from outer space!” To these people, the size of the supercomputer is the primary factor determining standing in the supercomputing world. They’ll talk to you if your supercomputer can be seen from low orbit, will feel sorry for you if your machine is only visible from low flying aircraft, and refuse to acknowledge your relevance if your system is only visible from the ground.

Now, in no way am I suggesting that the size of a supercomputer is not important. Almost self-evidently it is. A more powerful supercomputer can enable more realistic simulations, new kinds of science enquiry, or more comprehensive data analytics.

And sometimes real breakthroughs do occur as a direct result of the scale of the supercomputer used. But smaller supercomputers, even ones not visible from altitude, also deliver cutting edge science, engineering and data analytics.

2. The Apocalypse type. These people are convinced the end of supercomputing is at hand. Most of them focus on the technology challenges that seemingly erect impassable barriers to our progress. Exascale can’t happen because of power. And even if we could afford the power, we wouldn’t be able to program it. And even then, the system/application would collapse in a statistically inevitable heap of errors after a few minutes. Not to mention the skills shortage. We may as well give up now and just keep deploying the systems we have now.

Then there are the ones who proclaim that doom is not technological; it is political and financial. They argue that we cannot sustain the increasing budgets at the national lab scale, nor will senior managers in research-led businesses fund the increasing demand for supercomputer technology to enable higher fidelity simulation and deeper analytics. Quite rightly, they preach that simply quoting wonderful science is not enough justification (to the world outside of HPC) for these investments. This is where the Fort Knox types come in.

3. The Fort Knox type. It’s about the money to these people. They are driven by the big dollar deals, ideally, the high profit margin ones. They are reluctant to invest time in meetings, travel, projects, or acquisitions that don’t provide a substantial financial return this year. They often inhabit the parts of the ecosystem with better margins or with commercial applicability outside HPC (e.g. storage, networking, and so on) or sometimes can be found in those HPC vendors that keeping trying to discover a profitable business in selling solutions to buyers of technology with aggressive cost ambitions.

They are among the sharpest dressers, the most ardent advocates of their piece of the ecosystem, and get the least passionate interest from the buyers and users of supercomputers. Money and supercomputing always have been strange partners. Clearly, money, and lots of it, is required to fund the development, deployment, and operation of supercomputers and their enabling technologies (e.g. software).

But with such a technically dominated population of inhabitants, the HPC space often struggles to focus on this harsh reality. If companies can’t make money, or we can’t persuade politicians (and, by extension, the general public) to invest, then the R&D that is the lifeblood feeding the future of our world will weaken.

Likewise, the need to prove the economic return on investments in HPC services (both hardware and software) in commercial and academic/national lab spaces cannot be forgotten. The Fort Knox types have a crucial role to play in ensuring HPC continues to sustainably deliver its great potential. But the focus on money must balance with the pursuit of the technical race without which supercomputing would be meaningless.

4. The Art Gallery type. Keen to assure you as early as possible in the conversation that they are not technical – they leave technical detail to other people. Presenting a front of pride in their non-technical status, these people are usually, but not always, sales or business management people. However, a chink in their psyche appears, when they almost immediately follow up and make sure you know they once wrote some code.

Thus, they probably do recognize the integral value of technical understanding in the HPC world. But, either they are nervous about their lack of understanding (don’t be – not everyone is an expert – just be willing!) or they are hoping their stance will come across as above the detail (not good – this is often a detail game). And, in reality, the world of HPC is marked by significant technical expertise in so many of the people in sales positions, senior management or other traditionally business focused roles.

But there are also many who are not experts (or maybe aren’t anymore) but who have enough technology or science understanding to play their part in the ecosystem. Be proud of the technical knowledge you do have, honestly admit its limits, and be keen to learn more as needed. But then, you could say that about any skill, not just HPC.

5. The Horse-drawn Cart type.I remember when …” This person is able to turn any conversation about next year’s technology or this week’s implementation issue into a prolonged reminiscence of their distant childhood making supercomputers out of wooden sticks and spittle. Filled with “we tried that years ago,” “we had it much harder,” or “we should go back to the way we used to do it,” these monologues eventually stall as the polite but glazed expressions cemented on the faces around the room slowly reveal the audience has departed to mind-wandering land.

Occasionally, these reminiscences take a life of their own as dialogue springs forth – yes, those dreaded occasions when there is more than one Horse-drawn Cart type in the room. Sometimes, there are gems of insight relevant to the present or future to be found in these experiences though. The key to finding the gems is distinguishing the Horse-drawn Cart types from the Concorde types.

6. The Concorde type. Now that it is no longer flying the world with brutal performance and elegant class, this marvel of engineering brilliance and commercial application is distressingly easily seeping away from our memory. I’m writing this article sat in one of the other flagships of aviation, the Boeing 747, as I cross the Atlantic on my way to Salt Lake City for SC12 (there are about a dozen other HPC people just within a few rows of me).

But, much as I appreciate the 747 as probably the elder statesman of the skies, I wish the Concorde was still flying. Not that I’d expect to ride in it; it’s out of my league. But it is a shame that we have thrown away such a monumental capability: the 3 hour transatlantic crossing. That is a different class of interchange between London and New York than the current journey of a whole day’s flying.

More amazing still was that it was essentially 1960s technology. And this is my link to supercomputing. Some great technological achievements have peppered the history of supercomputing – processors, systems, algorithms, software implementations, etc. Many of them have been overtaken by subsequent products or technology shifts, but many we still rely on directly or via their evolutionary successors.  And, supercomputing too, is judged on the capability it enables, not merely the engineering brilliance of the technology implementation.

And the people part? Well, people made those great supercomputing technology advances. Some are sadly gone, many are still with us. In the fullness of the modern HPC ecosystem it is easy to let the impact of those technological leaps and their creators seep from our memory. Don’t.

So there you go – a selective stereotyping of the people that make supercomputing the marvel that we know – so powerful in its impact, often frustrating in its reality, usually addictive to those who encounter it, but always special. And, hopefully a few serious points about our community have been highlighted along the way. How many of these types did you see at SC12 this week? What types have I missed? Which, if any, are you?

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!

Quantum Watchers – Terrific Interview with Caltech’s John Preskill by CERN

July 17, 2024

In case you missed it, there's a fascinating interview with John Preskill, the prominent Caltech physicist and pioneering quantum computing researcher that was recently posted by CERN’s department of experimental physi Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to communities across the globe. As climate change is warming ocea Read more…

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical claims. A paper published on July 10 by researchers in the U. Read more…

Belt-Tightening in Store for Most Federal FY25 Science Budets

July 15, 2024

If it’s summer, it’s federal budgeting time, not to mention an election year as well. There’s an excellent summary of the curent state of FY25 efforts reported in AIP’s policy FYI: Science Policy News. Belt-tight Read more…

Peter Shor Wins IEEE 2025 Shannon Award

July 15, 2024

Peter Shor, the MIT mathematician whose ‘Shor’s algorithm’ sent shivers of fear through the encryption community and helped galvanize ongoing efforts to build quantum computers, has been named the 2025 winner of th Read more…

Weekly Wire Roundup: July 8-July 12, 2024

July 12, 2024

HPC news can get pretty sleepy in June and July, but this week saw a bump in activity midweek as Americans realized they still had work to do after the previous holiday weekend. The world outside the United States also s Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to com Read more…

Shutterstock 1886124835

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical c Read more…

Shutterstock 2203611339

NSF Issues Next Solicitation and More Detail on National Quantum Virtual Laboratory

July 10, 2024

After percolating for roughly a year, NSF has issued the next solicitation for the National Quantum Virtual Lab program — this one focused on design and imple Read more…

NCSA’s SEAS Team Keeps APACE of AlphaFold2

July 9, 2024

High-performance computing (HPC) can often be challenging for researchers to use because it requires expertise in working with large datasets, scaling the softw Read more…

Anders Jensen on Europe’s Plan for AI-optimized Supercomputers, Welcoming the UK, and More

July 8, 2024

The recent ISC24 conference in Hamburg showcased LUMI and other leadership-class supercomputers co-funded by the EuroHPC Joint Undertaking (JU), including three Read more…

Generative AI to Account for 1.5% of World’s Power Consumption by 2029

July 8, 2024

Generative AI will take on a larger chunk of the world's power consumption to keep up with the hefty hardware requirements to run applications. "AI chips repres Read more…

US Senators Propose $32 Billion in Annual AI Spending, but Critics Remain Unconvinced

July 5, 2024

Senate leader, Chuck Schumer, and three colleagues want the US government to spend at least $32 billion annually by 2026 for non-defense related AI systems.  T Read more…

Point and Click HPC: High-Performance Desktops

July 3, 2024

Recently, an interesting paper appeared on Arvix called Use Cases for High-Performance Research Desktops. To be clear, the term desktop in this context does not Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then 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_1687123447

Nvidia Economics: Make $5-$7 for Every $1 Spent on GPUs

June 30, 2024

Nvidia is saying that companies could make $5 to $7 for every $1 invested in GPUs over a four-year period. Customers are investing billions in new Nvidia hardwa Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

AMD Clears Up Messy GPU Roadmap, Upgrades Chips Annually

June 3, 2024

In the world of AI, there's a desperate search for an alternative to Nvidia's GPUs, and AMD is stepping up to the plate. AMD detailed its updated GPU roadmap, w Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very Read more…

Leading Solution Providers

Contributors

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l 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…

IonQ Plots Path to Commercial (Quantum) Advantage

July 2, 2024

IonQ, the trapped ion quantum computing specialist, delivered a progress report last week firming up 2024/25 product goals and reviewing its technology roadmap. 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…

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…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh 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…

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…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire