Skip to main content

2015 IBM Fellows


Jim Sexton

Jim Sexton

IBM Research
Department Manager, Data Centric Systems


Despite being a leader in IBM’s massive Blue Gene and CORAL computer systems projects, new IBM Fellow Jim Sexton got his start working with things considerably smaller.

“I’m a computational physicist at heart, and I’ve always been interested in how sub-nuclear particles interact. It’s the chemistry of the nucleus.”

After arriving in the United States from his native Ireland to earn his PhD in theoretical physics from Columbia University, Jim started work at Fermilab, a U.S. Department of Energy national laboratory specializing in high-energy particle physics.

It wasn’t until he completed a stint at the Institute for Advanced Study at Princeton University that he began to delve into the field of deep computing.

In 1988, Jim arrived in Yorktown Heights, NY at the Thomas J. Watson Research Center as a Research Fellow working on the GF-11 supercomputer with Monty Denneau and Don Weingarten. There he began a more holistic exploration of the field.

“The process I have been involved in since that time is called co-design. We assess and analyze applications performance on given systems, see where the bottlenecks were, and use the information to co-design the next generation of systems, software, and applications,” he said.

He took this experience with him back to Ireland, where he taught at his alma mater, Trinity College in Dublin, for 13 years. He returned to IBM Research and Yorktown Heights in 2004 as a research staff member on the Blue Gene team, and has progressed in the years since to lead the Computational Science Center within the Data Centric Systems group.

Jim was critically involved with all aspects of developing the revolutionary Blue Gene family of supercomputers, including chip verification and system bring-up. He was also responsible for the concept of collaborative application development with clients, which became a crucial part of the overall strategy for the commercial success of Blue Gene.

When Blue Gene was unveiled in 2004, it became both the most powerful and the most energy-efficient supercomputer in the world, consuming a fraction of the energy and floor space of any other system. The project launched IBM to the very forefront of the supercomputing industry, and the company was awarded the National Medal of Technology and Innovation by President Barack Obama in 2009.


It was the observation that to succeed, you have to commit everything you have. You have to invest all of your capabilities.


In his work, Jim has demonstrated the importance of application engagement throughout the lifecycle of the system. Within his team, he developed the techniques and methodology to understand, predict, and extract performance from these systems, used throughout the design process – from the initial stages through verification, test, bring-up, and acceptance.

It is this holistic approach that has driven the success and constant improvement of the supercomputing systems IBM has developed. As the field continued to evolve, however, computational improvements have grown more difficult to bring about.

“The big issues over the past 10 years have been how to continue to increase the productive power of computers,” he said. “Computers are getting more and more complex. It requires ever greater effort to increase their productivity, and in doing so, we’re driven to more and more complex architectures.

“We’re after greater computing power to drive advances in analytics, big data, and some of these other areas, but it’s getting harder and harder to achieve those.”

While he and his team continue to address those hardware challenges and innovate for the next generation of computing, Jim looks to what’s next in applications for these technologies. He is a thought leader in IBM’s analytics and big data performance analysis projects, and is a major driver of the current Data Centric System (DCS) architecture. He identified the emergence of workflows that require modeling, simulation, and complex analytics on massive data, which forms the basis of DCS, and co-developed the concept of computation near data that is essential to delivering systems solutions to address those workflows.

“We have opportunities for new applications of analytics and managing data,” he said. “It’s a very exciting time to be working with computers.”



Jim Sexton in his own words...


Who are some of the key mentors who helped get you to where you are today?

My original theses advisor Norman Christ was a hugely influential figure for me. He was just kicking off work on computer design for computational physics, and my start in that space came from him. While I was doing my post doc at IBM, Don Weingarten was important. He and Monty Denneau were putting together innovative systems that I was able to work on.

When I returned to IBM in 2004, Tilak Agerwala, Bill Pulleybank, and the whole Blue Gene team were outstanding to work with. Getting Blue Gene out the door together was an amazing time with an incredible group of scientists and individuals.


What do you look for in bringing new technologists to your team?

I’ve always liked physicists, but I’m probably biased. In all seriousness, I look for people who have a deep understanding of their domains – people who have passion, and can apply that passion to their work.


What was the best piece of advice you ever received?

It was the observation that to succeed, you have to commit everything you have. You have to invest all of your capabilities. You have to work hard. This isn’t surprising, but you look around at who has achieved great things, and you can immediately see their commitment, their work ethic, and their passion.


What was your first job, ever?

Pumping petrol at a filling station in Dublin.


What do you enjoy doing away from work?

I like to play computer games. For example, car racing games on my iPad, strategy games like Civilization or Gettysburg, things like that. I like to read history and science fiction. I also enjoy hiking and spending time outdoors.


What does it mean to you to be named an IBM Fellow?

It’s amazing; it’s very humbling. I look around IBM Research and I see all these incredible people. I very much believe that we succeed because of our people and those we collaborate with. I have had an incredible career here because of the people, and this is a very unexpected culmination of all that. I feel I’m hugely honored.



 

Watch the video

Watch how the Fellows program impacts IBM and the world

IBM Fellows by the numbers

This year's class

Year by year