recruiting and developing HPC staff


Lawrence Berkeley National Laboratory, USA


L


awrence Berkeley National Laboratory, a US Department of Energy (DOE) research facility located in Berkeley, California, operates one of DOE’s


premier supercomputing centres and conducts wide-ranging research in applied mathematics, computer science and computational science. Due to its proximity to San Francisco and Silicon Valley, Berkeley Lab also fi nds itself competing with both industry and academia when recruiting top talent. ‘Berkeley Lab is an exciting place to work,


with some of the world’s best scientists tackling critical societal challenges, such as global climate change, developing cleaner energy technologies, and advancing our basic knowledge across the wide spectrum of scientifi c disciplines,’ says Kathy Yelick, associate laboratory director for Computing Sciences at Berkeley Lab. ‘As a public institution, we can’t afford


the high salaries or stock options often associated with Silicon Valley, but we do offer


great benefi ts, and an opportunity to work with world-class scientists.’ Yelick adds that the Berkeley Lab’s ‘team


science’ environment, pioneered by namesake Ernest O. Lawrence in 1931, is a double-edged sword when it comes to recruiting. ‘Some computer scientists are inspired to be on a team working to cure cancer or understand the origins of the universe, but it means we are recruiting people who are willing to learn


focuses on depth in a particular area rather than interdisciplinary breadth.’ To fi nd the best candidates, computing


sciences recruiter Jeff Todd attends professional conferences and university job fairs. The computing organisation was one of the fi rst at Berkeley Lab to aggressively use social media for outreach and recruiting, activities that go hand-in-hand. Berkeley Lab’s location is also a plus. The Lab is located in the hills across the


BERKELEY LAB IS AN EXCITING PLACE TO WORK, WITH SOME OF THE WORLD’S BEST SCIENTISTS TACKLING CRITICAL SOCIETAL CHALLENGES, SUCH AS GLOBAL CLIMATE CHANGE


enough about biology, physics, chemistry or other fi elds that they can communicate with an interdisciplinary team.’ As one of the 250 Lab staff members who


also teach at the University of California at Berkeley, Yelick points out: ‘Berkeley has developed a graduate programme in Computational Science and Engineering to help train the kind of interdisciplinary researchers that we need to hire, but in general, the process of completing a PhD


bay from San Francisco with panoramic views of the Golden Gate Bridge. ‘With many potential employees, the


allure of California and the San Francisco region in particular are an easy way to begin the recruiting discussion,’ Todd says. ‘Factor in the Lab’s reputation, the quality of research being conducted and the overall working environment, and the result is often a persuasive package for hiring and retaining high-calibre researchers.’


Alessandro Curioni, manager, computational science, IBM Research – Zurich


engaging with students in fi elds such as computational chemistry, we can reach out to their colleagues and friends who may be studying mathematics systems, and other complementary subjects. HPC is an interdisciplinary fi eld and a holistic approach is required, as the skills required range from system design and software up to the analysis, creation and visualisation of the results. Telling someone that they should come


everything from hardware and integration to algorithms and application. One of the main things we need to convey is that by using supercomputing resources we are trying to solve real problems. We need to expose people as early as possible to the types of exciting problems that can eventually be solved with HPC, because it’s this interest that will attract people to our industry. The focus at the moment is very much on the application side, and by


W 38 SCIENTIFIC COMPUTING WORLD


hen talking about recruiting in HPC today, the issues that come up are diverse because it’s such a broad fi eld, covering


to work at an institution because it’s trying to develop the next generation of high- performance computer that will crunch far more numbers than its predecessors will not gain a lot of interest. But communicating the real problems that could be solved by building a machine of that type will. For the past few years we have participated in public fairs, in which we show our research to the larger public and demonstrate how HPC can be used to simulate an aircraft when taking off in an effort to reduce noise, for example, or how we can run simulations of the behaviour of a molecule. HPC is driving interest in disciplines like


mathematics, physics and chemistry, and we’ve noticed that the interest has grown at the same pace as the level of available


computational power – which has increased exponentially during the past 20 years. There is no large-scale scientifi c government project that does not include HPC-based simulation, and computational science has been accepted as the third pillar in scientifi c enterprise, together with theory and experimentation. At IBM we don’t have any project, even on the smallest scale, in which simulation is not making a difference. Recruiting people to do this type of


work is a challenge. Expertise is always required, but it’s not enough to be good in algorithmic development, for example; you need to know which is the best algorithm to use and how to implement it. And the ability to work in a


multidisciplinary team is also a must. Single units are a thing of the past as everyone is encouraged to talk to each other. As we move forward, the skills and experience required will become even more diverse, and working closely with universities will be crucial. The sector will be going through a transition soon – much like the time when digital computers were developed. It’s important to remember that the people we recruit now will be involved in forging the computing of the future.


www.scientific-computing.com


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48