InTheLoop | 08.16.2010
August 16, 2010
Dancing in the Dark: A New Look at Protein-Salt Interactions
Scientists using supercomputers at NERSC and x-ray absorption simulation software developed at Berkeley Lab’s Molecular Foundry are getting a new look at how proteins interact with simple salts in water, and what impacts these interactions may have on protein structures at the atomic level. Read more.
Web Photos That Reveal Secrets, Like Where You Live
According to a recent article in the New York Times, security experts and privacy advocates have recently begun warning about the potential dangers of geotags, which are embedded in photos and videos taken with GPS-equipped smartphones and digital cameras. Because the location data is not visible to the casual viewer, the concern is that many people may not realize it is there; and they could be compromising their privacy, if not their safety, when they post geotagged media online.
By looking at geotags and the text of posts, “you can easily find out where people live, what kind of things they have in their house and also when they are going to be away,” says Robin Sommer of the Advanced Computing for Science Department in CRD, who works with the Berkeley Lab cybersecurity group. Sommer and Gerald Friedland, a colleague at the International Computer Science Institute at UC Berkeley, presented their paper “Cybercasing the Joint: On the Privacy Implications of Geotagging” last week at the Fifth USENIX Workshop on Hot Topics in Security (HotSec 10) in Washington, D.C.
Top 10 Near Miss and Safety Suggestions Are Due August 31
Berkeley Lab Computing Sciences staff are reminded to send near miss safety hazards and safety suggestions to Safety Coordinator Betsy MacGowan. Horst Simon, Betsy MacGowan and the Safety Committee will select the CS Safety Top 10 suggestions that provide the greatest improvement to Lab safety. Employees who submit the winning entries will be recognized with a Spot Award, which includes a cash prize, and Top 10 winners will be featured on the CS website.
This safety competition has been launched as part of the larger initiative to transform the safety culture at the Lab. The competition is open to CS staff, guests and matrixed employees, and is intended to increase awareness of safety issues within CS, CRD and NERSC. All submissions for the first cycle are due on August 31, and the winners will be announced in September. Cycle deadlines, winners and the Top 10 list will be announced periodically in InTheLoop.
ACTS Workshop Will Be Held This Week on UC Campus
The 11th Workshop on the DOE Advanced Computational Software (ACTS) Collection, “High Performance Software Tools to Fast-Track the Development of Scalable and Sustainable Applications,” will be held in Sutardja Dai Hall on the UC Berkeley campus this Tuesday through Thursday, August 17-20.
The four-day workshop, organized by Tony Drummond of CRD with logistical help from Yeen Mankin, will present an introduction to the DOE ACTS Collection for application scientists whose research demands include either large amounts of computation, the use of robust numerical algorithms, or combinations of these. The workshop will include a range of tutorials on the tools currently available in the collection, discussion sessions aimed to solve specific computational needs by the workshop participants, and hands-on practice using state-of-the-art supercomputers at NERSC.
Associate Lab Director Horst Simon will give a keynote address on current and future trends in high performance computing (HPC). Sherry Li of CRD will give a tutorial on SuperLU. And Drummond will speak on the DOE ACTS Collection and fast-tracking the development of HPC applications.
Python Boot Camp Scheduled for August 23–25
Registration is now open for the Python Boot Camp to be held on the UC Berkeley campus August 23–25. The purpose of the boot camp is to get those familiar with other computing languages (like C, Java, Fortran, and Lisp) ramped on the basics of the Python language. The Boot Camp itself is a mixture of formal lectures, in-class demos, coding breakout sessions for participants, and homework projects. Registration is open to anyone in the UC Berkeley research/academic community on a first-come basis.
CITRIS CSE to Open Visual-Analytic Lab
The CITRIS Computational Science and Engineering program is in the final stages of setting up a Visual-Analytic Lab at CITRIS. If you are interested in having access to this lab’s resources, please contact Masoud Nikravesh by September 3, 2010.
SDSC to Host “Data-Intensive Discovery” Conference October 26–28
The San Diego Supercomputer Center (SDSC) will host a special conference as it prepares to deploy a data-intensive, flash memory-based high-performance computing (HPC) system called Gordon in mid-2011. Called “Grand Challenges in Data-Intensive Discovery,” the multidisciplinary conference will be held October 26–28, 2010 at SDSC, located on the UC San Diego campus. The program includes Peter Nugent of NERSC and CRD speaking on “Transient Sky Surveys.”
This Week’s Computing Sciences Seminars
Pab Lab Boot Camp—Short Course on Parallel Programming
Monday, August 16, 9:00 am–6:00 pm, CITRIS Auditorium, UC Berkeley
Tuesday, August 17, 8:45 am–6:00 pm, CITRIS Auditorium, UC Berkeley
Wednesday, August 18, 8:45 am–5:00 pm, CITRIS Auditorium, UC Berkeley
The 2010 Pab Lab Boot Camp—Short Course on Parallel Programming is intended to offer programmers a practical introduction to parallel programming techniques and tools on current parallel computers, emphasizing multicore and manycore computers. Three all-day sessions will be held at CITRIS Auditorium on the UC Berkeley campus, as well as online. The course is free to students, faculty, staff, and affiliates (including Berkeley Lab), but registration is required.
Tuning Parallel Applications in Parallel
Thursday, August 19, 10:00–11:00 am, 50F-1647
Ananta Tiwari, University of Maryland
In this talk, we will present Parallel Active Harmony — a scalable, end-to-end auto-tuner for scientific codes. Harmony takes a search-based collaborative approach to auto-tuning. Application programmers and end-users collaborate to describe and export a set of performance related parameters to the Harmony system. These parameters define a tuning search-space. The auto-tuner monitors the program performance and suggests application adaptation decisions. The decisions are made by a central controller using a parallel search algorithm. The parallel search algorithm leverages parallel architectures to search across a set of optimization parameter values. Different nodes of a parallel system evaluate different configurations at each timestep.
Recently, we have been working on making online tuning practical. Our goal is to enable application developers to write applications once and have our auto-tuner adjust the application behavior automatically when run on new systems. For tunable parameters that require new code (for example, unroll factors), the Harmony system generates code on-the-fly using third-party code-generation tools. Effectively, this merges traditional feedback-directed optimization and just-in-time compilation. We evaluated our system on a parallel multi-block lattice Boltzmann code. The code has five hot regions, and all five are tuned simultaneously at runtime. Performance improved up to 48% for a 512-core run with a minimal cost of 2% (8 cores) extra cores for code-generation.
Link of the Week: Video Game Tackles Protein Folding
In a match that pitted video game players against the best known computer program designed for the task, the gamers outperformed the software in figuring out how 10 proteins fold into their three-dimensional configurations, according to John Markoff in the New York Times. In a comparison involving 10 separate protein-folding puzzles, video game players matched the results generated by software solutions in three of the puzzles, outperformed them in five cases and found significantly better solutions in two others.
About Computing Sciences at Berkeley Lab
The Lawrence Berkeley National Laboratory (Berkeley Lab) Computing Sciences organization provides the computing and networking resources and expertise critical to advancing the Department of Energy's research missions: developing new energy sources, improving energy efficiency, developing new materials and increasing our understanding of ourselves, our world and our universe.
ESnet, the Energy Sciences Network, provides the high-bandwidth, reliable connections that link scientists at 40 DOE research sites to each other and to experimental facilities and supercomputing centers around the country. The National Energy Research Scientific Computing Center (NERSC) powers the discoveries of 6,000 scientists at national laboratories and universities, including those at Berkeley Lab's Computational Research Division (CRD). CRD conducts research and development in mathematical modeling and simulation, algorithm design, data storage, management and analysis, computer system architecture and high-performance software implementation. NERSC and ESnet are DOE Office of Science User Facilities.
Lawrence Berkeley National Laboratory addresses the world's most urgent scientific challenges by advancing sustainable energy, protecting human health, creating new materials, and revealing the origin and fate of the universe. Founded in 1931, Berkeley Lab's scientific expertise has been recognized with 13 Nobel prizes. The University of California manages Berkeley Lab for the DOE’s Office of Science.
DOE’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.