October 12, 2007
Dr. Valentina Salapura is a computer architect in the Exploratory Server Systems group at the IBM T.J. Watson Research Center, where she is helping to define the next generation of computer systems based on chip multiprocessors and parallel processing. She has studied the use of multiprocessor systems in a range of environments, from automotive applications to network processors and supercomputers.
On October 18 at the Grace Hopper Celebration of Women in Computing Conference, she will be delivering a plenary speech titled Next Generation Supercomputers. We got the opportunity to ask her about her upcoming talk and the significance of the conference to women in the field of computer science.
HPCwire: Tell us about the Grace Hopper Conference and why it is important for women?
Salapura: The Grace Hopper Celebration of Women in Computing is an event to bring together the women in the field and to mentor students and colleagues early in their computer science and engineering careers. Like many science and engineering disciplines, it is very male dominated, and it is easy for somebody starting out to feel lonely, isolated and discouraged.
It's important to build networks for women, to help each other out. I attended a recent talk by Prof. Rosser, the Dean of the Georgia Tech, about challenges women face in science and engineering today, and their unequal position compared to male colleagues. Prof. Rosser found that a woman needs 2.5 times more achievements compared to her male colleagues to reach the same level of recognition. That's why it's so important that we support each other and build our own networks.
Take as an example this year's Turing Award winner. My dear colleague and friend Fran Allen was an early pioneer of parallel programming and she received the Turing Award this year. She was the first woman in history to receive it, after 40 men. Fran was also the first IBM Fellow, and the first female president of the IBM academy, our internal think tank. It's been long overdue that a woman is recognized in this way.
HPCwire: Can you tell us more about women pioneers in the field of computer science?
Salapura: There are not many women in computer science. In addition to Fran Allan, whom I already talked about, two names stand out: Grace Hopper and Anita Borg.
Grace Hopper was another one of the early pioneers of computing field. She was behind the definition of a programming language that powers the systems in many banks and financial systems. The conference is named in her honor. It is now becoming a yearly event, and growing rapidly. Last year the event had over 1,200 attendees.
Another key figure, especially for women in our field, was Anita Borg. Anita Borg was one a few women PhDs in our field when she graduated from NYU. She made important contributions to system design in her early career at various companies. But her impact goes beyond her technical work. She recognized early on the need attract and advance women in technology. Anita Borg founded the Institute for Women and Technology. She died in 2003. Today the institute carries her name and continues the work. The institute is the organizer of the Grace Hopper Conference.
HPCwire: Tell us about your participation at the conference.
Salapura: I have been involved with the Grace Hopper Conference in the past, helping to organize IBM's participation. IBM puts great emphasis on workforce diversity and support for underrepresented minorities, and we're proud to be a longtime sponsor of the conference. Ultimately, encouraging women participation is the best way to ensure that women are adequately represented in the workforce. This year, IBM will be a Platinum Sponsor and we'll be hosting a reception for the conference on Friday night, which I'm helping to organize.
HPCwire: I gather you are invited to give a plenary speech at the Grace Hopper Conference? What are you talking about?
Salapura: I feel greatly honored by this invitation to give a plenary speech. I'll be talking about my technical field, computer architecture. As you may know, computer architecture is going through a major industry-wide revolution right now. From faster and faster single processors we're shifting to multiprocessor systems, where multiple processors share in the work. I have worked on multiprocessors for a long time, so this shift is like a personal victory.
Specifically, I'll be talking about the next generation of the Blue Gene system. As you may know, Blue Gene is IBM's top ranked supercomputer and has been the world's fastest system for several years. This new system means a lot to me personally, since I contributed to virtually every part of the system. I served as the lead for the memory coherence architecture with snoop filters, which ensures that the processors can share data correctly with low overhead, and was also the lead for the performance monitoring unit, which keeps track of how well the computer is working. In addition, I led the bringup work to assemble the initial system prototype once our chip came back from manufacturing, coordinating a large cross site team.
HPCwire: So what's new with supercomputers?
Salapura: It has changed drastically! In the past, people did not care about power and energy consumption, and the focus was to deliver performance at any price. But computers have become so powerful, and in the process started using so much energy. For example, they needed to construct a special building with special transmission lines for the Earth Simulator supercomputer, which was built in Japan. That was a very impressive computer, which held the number one spot for two years.
To top that system, it was necessary to fundamentally re-think supercomputer design. To build bigger and faster computers, we have to build computers that use less power so we can put more components in the datacenter, the building that houses the computer. Because the computers are generating so much heat, we need to limit the power they consume and hence the heat they generate. Only half of a datacenter's electricity bill goes to powering the computers. The other half goes for the air conditioning, or the computers would overheat and be damaged.
To show the impact of this, there's even a new list of top computers. The old list was called Top500, the 500 fastest supercomputers on the planet. The new list will be called Green500, the 500 most energy efficient computers on the planet!
Dr. Salapura has been a technical leader for the Blue Gene program since its inception at the IBM T.J. Watson Research Center. She has contributed to the architecture and implementation of several generations of Blue Gene Systems focusing on multiprocessor interconnect and synchronization and multithreaded, multicore architecture design and evaluation. Most recently, she has been unit lead for several units of Blue Gene/P, as well as a leader of the chip and system bringup effort.
Before joining IBM Research at the IBM T.J. Watson Research Center in 2000, Dr. Salapura was Assistant Professor with the Dept. of Computer Engineering at Technische Universität Wien in Vienna, Austria. Dr. Salapura received a PhD degree in Computer Science from Technische Universität Wien in Vienna, Austria, and MS degrees in Computer Science and Electrical Engineering from the University of Zagreb, Croatia.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?