Faces of HPC: John West

Tell a bit about yourself.

I was born in the US and moved around a bunch when I was a kid...including some time in Europe. We settled down in Mississippi in the early 1980s, and I've been here ever since (readers may be surprised to learn that Mississippi has a strong role in HPC; when I was tracking rankings back in the early 2000s Mississippi routinely ranked in the top 10 US states by HPC capability). I went to school at Mississippi State University and got a BS in Electrical Engineering followed by an MS in Computational Engineering. These days I'm one of the few people in my workplace at the University of Texas without a PhD - I did make it a fair way down that road, but life got in the way so I'm ABD right now, and probably always will be.

Outside of work, I'm mostly a dad to Emma (7), Jacob (13), John David (15), and a partner to Bobbie, my wife and best friend. Sometimes I get a chance to read a book or two, mostly sci-fi and fantasy, with the odd book about typography and the history of printing and bookbinding thrown in for good measure.

What is your current job?

I'm currently at the Texas Advanced Computing Center (www.tacc.utexas.edu), part of The University of Texas at Austin. At TACC I'm the Director of Strategic Initiatives, which is a fancy way of saying I do a little bit of everything but mostly help provide extra support as TACC pushes into new technologies or lines of business. One of the things that attracted me to TACC is their commitment to the HPC community, and to leaving the world a little better than they found it. They have been very supportive of my involvement in activities such as the annual SC conference (for which I was general chair in 2016) and the ACM Special Interest Group in HPC (where I serve as vice chair). TACC is also a leader in efforts to attract and retain women and underrepresented minorities in computing and HPC, an issue that has become more important to me over the years.

How did you become interested in HPC?

I got into HPC very accidentally. When I finished my BS'EE in 1992 the job market for engineers in the US was terrible, and it didn't help that I was a horrible job applicant who didn't know what he wanted to be when he grew up. One of my professors suggested I should consider a graduate degree, which I did mostly to keep from living in my car. But it was a great decision! I was lucky enough to be part of the National Science Foundation Engineering Research Center for Computational Field Simulation that had been recently established by Joe Thompson and his colleagues at Mississippi State. Over the years Professor Thompson was an incredible mentor and example for me, and working in the environment he created was...amazing. Especially looking back as a grown up, I realize how fortunate I was to be in that place at that time. It was there that I first saw how useful computers could be in solving real problems, and I was hooked. The ERC had a small Intel iPSC/860, and thanks to some very enthusiastic professors I was able to do really exciting work early on.

While I did actual technical work in the early part of my career, I got tracked into HPC program management fairly early on. This is probably a consequence of being comfortable with public speaking early in my career and a quick study in management, but only a middling technical contributor (if I’m honest). I started my career as an employee of the Department of Defense working in one of their HPC centers in Mississippi. Over the years I moved between contractor and government roles to pursue interesting opportunities, but always associated with the same HPC center. I eventually ended up leading the program that funded that center -- the DoD's High Performance Computing Modernization Program. I was director of that program for four years and, because of its scope (the HPCMP is a nation-wide program with 5 HPC centers, about 700 employees, and a $300M annual budget), I had a tremendous opportunity to learn a wide variety of new things, from the way the US legislative and executive branches work together, to the ins and outs of large-scale contract management.

As part of this project we want to celebrate the diversity of HPC, in particular to promote equality across the nine “protected” characteristics of the UK Equality Act, which are replicated in world-wide equality legislation. Do you feel an affiliation with this matter, and if so how has this interacted with or impacted your job in the HPC community?’

This is an issue that is important to me, and I've been fortunate over the past several years to be in positions of leadership in organizations that gave me the freedom and tools to try and help expand the diversity of the HPC workforce. My motivation is engineering-oriented: HPC helps scientists and engineers make the world a better place, but building technologies that are effective in solving science problems is really hard. To build great solutions we need to ask all the smart people for their best ideas. Today as a community we aren't doing that, and so we cannot be sure we as HPC professionals are delivering the best tools we can build. I'd like to be part of changing that. Unfortunately this is one area in which we do not yet know how to make rapid progress; the evidence we have to date is it will take decades of consistent, methodical, incremental effort to alter the demographics of HPC (and computing in general).

Is there something about you that’s given you a unique or creative approach to what you do?

I'm not sure there is anything special about me, but I do think that it has been very useful for me to be able to admit when I didn't know something, and to ask questions. Early in my career I think people interpreted this behavior as enthusiastic curiosity, and later in my career as a leader of a large organization people seemed to appreciate being asked to share their expertise as part of a group decision-making process. While both things are probably true, in reality this was mostly about me being focused on getting to the best answer we (as a group) could find, no matter who eventually got the credit. Harry Truman is credited with the saying "It is amazing what you can accomplish if you do not care who gets the credit." For me this philosophy is a natural way to create a healthy organization where everyone has the chance to contribute what they do best.

Were there any challenges when you first entered the field?

The thing that is most attractive to me about HPC as a profession is that it isn't really one field: it's many disciplines, from computer graphics and visualization to programming to management and policy making, all put together. This means that its possible to be an HPC professional for 20 or 30 years and still regularly try new things and find new things to learn. Many of the colleagues I worked with at the beginning of my career had been studying their specific topic -- for example, code optimization -- for decades, and were amazing technical contributors. I discovered early on that my own most productive work style is as a generalist, with a broad understanding of many different areas. At the start of my career I was challenged to reconcile the deep specialist model of career development, which I thought was the only right model, with my own preferences. Eventually I discovered that my generalist tendencies could be a strength, but I wasted many years trying to force myself into the specialist mold that (it seemed) everyone around me fit so easily into.

A very successful senior leader told me about 7 years into my career that eventually I would have to "grow up" and stop trying new things all the time. Although he taught me many helpful things, happily it turns out that he was wrong on that point.

What’s the best thing about working in HPC?

As a field for me the best thing has been the wide diversity of topics that all go together to create our discipline. A side effect of that is that the people who work in HPC centers have wonderfully curious personalities and varied interests, and it makes these centers the very best place I can imagine working.

If there’s one thing about HPC you could change, what would it be?

When I started my career in the early 1990s, the HPC hardware and software landscape was much more diverse. Companies were trying radically new approaches to parallel computing every year...I miss that diversity and experimentation. I understand why the field today is more commoditized, and I recognize that standardization has brought HPC to a much wider group of users, but I still miss the variety of cool machines we used to have.

What’s next for you in HPC?

I've never had a career plan beyond "find a way to be helpful while learning new things". So, who knows what's next?

John West was interviewed by Toni Collis in March 2017.