Category: Argonne Leadership Computing Facility

Categories
Argonne Leadership Computing Facility

A unique postdoc opportunity in HPC

butler-fellowship.blogFebruary 28 is the deadline to apply for the ALCF’s new computational science postdoctoral fellowship position, named in honor of the late Argonne mathematician and computer science pioneer Margaret Butler.
From very early on, Margaret viewed computers as a technology with great potential to push the frontiers of science. Over the course of her career, she also worked to establish the careers of many other women in computing. She understood the value of promoting change by creating opportunity. And so it was especially apt that this new fellowship commemorating her contributions to the field would be a unique career opportunity.
This fellowship, which is open to both men and women, is an opportunity for the recipient to work with Argonne scientists in support of scientific discovery in their field of expertise. It’s an opportunity to work in a multidisciplinary environment with world leading experts in computational mathematics and computer science, and to use some of the world’s most powerful computer systems. Mostly, it’s an opportunity for a bright, young researcher to launch a computational science career in pursuit of significant achievements in science.
More information about The Margaret Butler Fellowship in Computational Science can be found online.

Categories
Argonne Leadership Computing Facility

New Year, New INCITE, and New Insight

INCITEFor most of us, New Year’s Day marks the start of new opportunities and renewed potential. Fittingly, January 1 also begins a new cycle of INCITE projects, and 2014 is the year that will see the highest number of awards and core hours ever granted by the DOE Leadership Computing program in its 10 year history. Over the next 12 months, 59 teams of researchers will share nearly 6 billion core hours of peer-reviewed parcels of supercomputing time at Argonne and Oak Ridge national laboratories to work on energy and environmental problems with the potential to benefit the world.
INCITE investigations are the largest and hardest problems in science and engineering, often slow and uncertain by their very nature, and are assessed using a wide range of metrics. The advancements that come out of INCITE work are vastly accelerated with each new generation of computing system, and help to gain a deeper understanding of highly complex systems in our physical world. These projects build knowledge through bigger and better simulations, improved algorithms, and novel computational techniques. Demystifying the world in this manner requires copious amounts of preparation and processor hours, and each new year of INCITE awards brings a renewed sense of what’s possible. To track this year’s INCITE discoveries at Argonne, I invite you to visit the Argonne Leadership Computing Facility website.

Categories
Argonne Leadership Computing Facility Research

Beyond the Standard Model

utfit-fullThe discovery last year at CERN of the Higgs boson — a particle that may well be responsible for all the mass in the universe — was momentous to physicists everywhere. The revelation of Higgs is critical to validating a nearly five-decade-old fundamental physics theory, known as the Standard Model, which accounts for all known subatomic particles and their interactions. Scientists, meanwhile, continue their search for answers to weighty unexplained physical phenomena such as the existence of dark matter and what happened to all the antimatter since the Big Bang.
Fermilab theoretical physicist Paul Mackenzie is leading a multiyear project at the ALCF to shed light on the mysterious particles and forces associated with “physics beyond the Standard Model.” According to Mackenzie, the Standard Model has many complex and peculiar features that have led to the nearly universal belief that there is new, as yet undiscovered physics which will explain these features.
Mackenzie heads a national effort to leverage HPC resources to advance quantum chromodynamics (QDC), the study of how quarks and gluons interact. Supercomputers like Mira enable scientists to study quarks and gluons in situations that are not possible in accelerator and cosmic ray experiments, and have the computational power needed to give quark-antiquark pairs their proper, very light masses for the first time — removing one of the largest remaining uncertainties involved in QCD calculations. Read more about Mackenzie’s research at the ALCF here.

Categories
Argonne Leadership Computing Facility

Strong ALCF showing at SC13

Habib_SC13_blogThe highly competitive technical program of the annual Supercomputing conference shows broad participation from ALCF researchers this year. Several papers coauthored by ALCF researchers were accepted — one of which is also a finalist for the ACM Gordon Bell Prize — and all feature work either performed on or optimized for Argonne’s Mira supercomputer. Other ALCF participation runs the gamut, from posters and workshops to broader engagement and round table sessions.
Fifteen DOE National Laboratories, including Argonne, will be represented in the exhibit hall in one booth (1327) under the theme of “DOE: HPC for a Greener, Smarter, Safer World” and host presentations, electronic posters, 3D simulations, demonstrations and roundtable discussions.
I’ll be there, trying to take in as many sessions as possible and meeting old friends, colleagues and collaborators. If you want to connect, drop me a line at [email protected]
Two papers that I’m involved with are being presented by their main authors. Here are the links:
Integrating Dynamic Pricing of Electricity into Energy Aware Scheduling for HPC Systems

Performance Characterization and Prediction Based Modeling of Collective Two-Phase I/O

Categories
Argonne Leadership Computing Facility Research

Accelerating the discovery of alternative fuel sources

In many ways, biofuel research is like modern day alchemy. The transmutation of biomass materials — which includes anything from kitchen and latrine waste to stalky, non-edible plants — into a sustainable and renewable energy source involves catalysts and chemical reactions. The process promises to help meet the world’s critical energy challenges.
Biofuel research can also be thought of as the ultimate multi-scale, multi-physics research problem. It represents several interesting biological supply-chain management problems. Not surprisingly, biofuel research spans several domains here at Argonne, and takes place in wet labs and joint institutes across the lab campus. There is also an exciting INCITE research project going on in the ALCF aimed at finding a more effective way to convert plant materials that contain cellulose, such as wood chips and switchgrass, into sugars, and then converted into biofuels.
A science team from the National Renewable Energy Laboratory is using Mira to conduct large-scale simulations of the complex cellulose-to-sugar conversion process. Researchers are able to obtain data, such as the level of an enzyme’s binding free energy, which is difficult to obtain through conventional experimental approaches, helping to accelerate the process of screening and testing new enzymes. With such information, researchers will be able to identify potential enzyme modifications and then feed their discoveries into experiments aimed at developing and validating improved catalysts. Read the full research highlight here.

Categories
Argonne Leadership Computing Facility Education

Extreme-Scale Computing Training Course: Class of 2013!

Today I met with the first class of young scientists and researchers to participate in the Argonne Training Program on Extreme-Scale Computing. The trainees are now into week two of lectures and hands-on sessions aimed at teaching them how to program massively parallel supercomputers. I chaired the afternoon session on data visualization and analysis, complete with a set up success stories, taught by colleagues from ALCF, the University of Oregon/Lawrence Berkeley National Laboratory, and Kitware, Inc.
Training course organizers tell me this is an enthusiastic and motivated group; many of the participants remain long after the lectures end to engage the speakers on topics ranging from the latest performance tools to debugging to data analysis. Last week the group got special access to Argonne’s leadership computing resources, including Mira. This week they got a similar opportunity to experiment with application runs on Oak Ridge National Laboratory’s leadership system, Titan.
By funding this training course, the DOE is helping expand the user community of today’s high-end systems, but more importantly they are helping prepare a new generation of computer and computational scientists to keep our national priorities on track.

Categories
Argonne Leadership Computing Facility

Expanding the community, accelerating mission-critical research

Summertime, specifically July 1, is when the ASCR Leadership Computing Challenge (ALCC) projects get underway at the Leadership Computing Facilities at Argonne and Oak Ridge, and at the National Energy Research Scientific Computing Center (NERSC). The supercomputing centers will support a total of 32 projects and 1.6 billion core-hours — 809 million core-hours to 13 new projects at the ALCF alone.
ALCC projects expand into new areas of science and engineering of interest to the DOE mission — the “high-risk, high-payoff” research aimed at, among other things, national emergency mitigation — and also serves to grow a critical demographic: the community of researchers capable of using leadership computing resources.
This is the first year that ALCC projects will gain access to Argonne’s Mira system, which will greatly accelerate the target research in clean energy, climate change prediction, and battery research. More information about the individual 2013 ALCC awards can be found here.

Categories
Argonne Leadership Computing Facility Research

Cracking the source of crackle in supersonic jet noise

The ASCR Leadership Computing Challenge projects that make up roughly 30% of the time awarded on ALCF supercomputers each year go to support “high-risk, high-payoff” simulations of interest to the DOE. Stanford’s Parviz Moin used his 60 million hour award to make a new and potentially industry-changing discovery about the source of supersonic jet noise. No ear protection needed.
Moin and his team ran large eddy simulations on Intrepid to determine the source of crackle in hot supersonic jet engines. Crackle is a major source of engine noise that causes hearing damage and impacts fuel efficiency. This particularly irritating phenomenon is associated with shock-like “N-shaped” acoustic waveforms consisting of sudden strong compressions followed by more gradual expansions. Because crackle occurs in the direction of peak jet noise, its elimination has the potential to help meet the U.S. Navy’s near-term jet noise reduction goal of 3 dB in the peak noise.
One way to make jets less noisy is to modify the shapes of the engine exhaust nozzles using pointed cutouts, called chevrons. Past ALCF allocations to Moin enabled a comprehensive study of the physical mechanisms by which chevrons affect the jet mixing and shock-associated noise. His current allocation was used to complete the simulations capturing crackle events and to develop new methods to identify and save such events for further study. Furthermore, with the source of the crackle noise now identified, new nozzle designs can be simulated.

Categories
Argonne Leadership Computing Facility Technology

Renewed urgency in the race to exascale

At a special two-day symposium last month, Argonne invited back many of the visionaries who in some way contributed to the lab’s 30 years in advancing parallel computing and computational science. Most if not all of these individuals have made a career in high performance computing — directing programs and conducting research in government agencies, national laboratories, universities and industry. Strong DOE investment in research and development of HPC capabilities has kept the U.S. the leader in high impact scientific research for decades, but now the DOE’s shining scientific supercomputing centers are oversubscribed by factors of three or more. Exascale is the next and necessary milestone in the race for computing power that will enable future breakthroughs in energy, medicine, and engineering, and the U.S. is now facing fierce competition from around the world.
Argonne’s Rick Stevens has been a leader in DOE planning for exascale since 2007. He recently testified to the urgent need for sustained government investment in exascale at a Congressional Subcommittee on Energy hearing, “America’s Next Generation Supercomputer: The Exascale Challenge.” The May 22 hearing was related to a bill proposed by Illinois Rep. Randy Hultgren to improve the HPC research program of the DOE and make a renewed push for exascale research in the U.S. A full transcript of his testimony can be found here.

Categories
Argonne Leadership Computing Facility

March 14 Argonne OutLoud Public Lecture

Research teams from around the world are preparing to use Mira, Argonne’s new petaflops supercomputer, to tackle the most challenging problems in science and engineering today. What happens when a star explodes? Can we find alternative fuel sources? How hot will the greenhouse world be? These problems cannot be addressed any other way because of their sheer size or complexity. Mira will be used to compress time and explore many possible solutions to issues related to energy and the environment. It will enable the prototyping and testing of construction materials before they are ever physically built and will help search for effective vaccines to deadly infections.
On March 14, Argonne’s Pete Beckman will give a talk about how math and supercomputers are accelerating scientific discovery as part of Argonne’s public lecture series Argonne OutLoud. Dr. Beckman is the co-director of the Northwestern-Argonne Institute for Science and Engineering.