Recognizing the rock stars of HPC

Print Friendly, PDF & Email

Last month I posted a note about the upcoming July 1 deadline for the Cray, Fernbach, and Kennedy awards following on from a post on Dan Reed’s blog. This started a conversation between Dan and I about the awards and their importance in our community.

It seems to me that what we really need in HPC is more rock stars. People that are well known enough within our community that they serve as visible entry points for the outside community and glue us together. When I started my career there were several of these folks around, and they were very well known. People like Larry Smarr, Ken Kennedy, Joe Thompson, and others who were changing the face of high performance and scientific computing and recognized by the outside world.

We still have these people, engineering dramatic change and creating the future, but our community doesn’t put a focus on recognizing these special people as, well, special.

HPC, at least at the high end, is still a very niche profession. People who achieve dramatic results in our community don’t (usually) get rich, and they don’t get an invite to the Oscars. They are paid in recognition, and this in turn has the salutary effect of motivating the youngest members of our community to bust ass. And the inevitable feature stories in mass media when something really exciting happens gives our community the chance to attract new blood. I still remember stumbling over myself when, as a new Master’s student in computational engineering, I was introduced to Larry Smarr who at the time was doing pioneering work in clusters at NCSA. I ended up working in or with the organizations that Dr. Smarr, Dr. Kennedy, and Dr. Thompson led for several years, and that time had a profound impact on my career (of course, you’d need to ask my bosses whether it actually made a difference to them, and I’m not giving out email addresses).

All of which brings us back to the big HPC awards, each stewarded by the SCxy committee. There are three: the Seymour Cray Award, the Sidney Fernbach Award, and the Ken Kennedy Award.

The Cray award honors outstanding HPC systems architects; past winners include Steve Wallach (Convex, Convey), Tadashi Watanabe (NEC SX), Steve Scott (Cray T3E, X1), Burton Smith (of Tera fame), and others.

The Fernbach award recognizes the outstanding application of HPC using innovative approaches. Past winners include Bill Gropp (MPI), Jack Dongarra (LAPACK, etc.) and many others.

The newest award, announced this year, grows the Big Two to the Big Three, and honors Ken Kennedy’s achievements by recognizing those who have made significant contributions to programmability and productivity in computing.

To evaluate my rock star theory of the value of awards in our community I reached out to last year’s recipient of the Cray award, Steve Wallach, and asked him what the award meant to him. Did it matter, or was it just one more piece of schwag he never looks at?

This is what he said

First and primary, personally I do not seek awards. This is important to know. I just do my thing and try to do a good job as an engineer. Getting an award, especially one that comes from my peers, is special. Peers determine how effective your output is. Recognition by your peers is extremely important feedback, and to be candid, is one of the few things that affect me.

Using a car analogy, when a new car is introduced, a review acknowledging that this is as good as a Mercedes (or Porsche in my case) is worth more than almost any other comment. Comparing to a known “gold” standard is very succinct.

In his view, receiving the Cray award meets both of those criteria — an award unsought, given unexpectedly by his peers, and named after the gold standard in supercomputing, who’s work Steve has studied throughout his own career.

And what a career: 33 patents; member of the National Academy of Engineering; IEEE Fellow; founding member of PITAC (The Presidential Information Technology Advisory Committee); co-founder of Convex; a career in venture capital; Vice President of technology for Chiaro Networks; and now founder of HPC’s bouncing new baby supercomputing company, Convey Computer. In short, a rock star.

When I heard that I received the Cray Award it left me speechless. Anyone who knows me has never seen me speechless. That was a sincere reaction. In my 40 year career, the Cray Award, is the most significant technical milestone.

Steve tells me that he has a shelf at home with many of his awards; pieces of plastic encasing the public offering of Convex, for example, and a plastic memento of the HP purchase of that company. All of this was moved off the shelf to make room for the Cray award, where it stands by itself today as a testament to a long, successful career.

These awards matter to our rock stars. It is worth your time and your serious thought to nominate our community’s best for these awards each year. Not only to pay back those who have given so much to our community and to the world, but also to pay forward as we try to inspire the best and brightest of a new generation.

Trackbacks

  1. […] written before on this site about the tremendous impact these awards can have on their recipients. In that article I wrote […]

Comments

  1. John Leidel says

    I definitely second this motion. As a young and eager engineer, I was allowed to work under Jim Ballew. Jim designed countless HPC systems over the years. He worked side by side with great people from Cray, Silicon Graphics, Convex and Compaq. Most recently, he was CTO at Appro. I learned more from Jim in six months than I did in several years of undergraduate coursework. It was absolutely a pleasure to work him.

    I encourage all young engineers to find a capable mentor. Aside from guidance, they provide you a creative outlet for which to vent your thoughts.

  2. Well said indeed. Thank you for writing this article and bringing the importance of these awards to our community.