A-Z Index | Phone Book | Careers

InTheLoop | 02.22.2011

February 22, 2011

LBNL-Tsukuba Joint Workshop This Week

The Computing Sciences Directorate at Berkeley Lab and the Center for Computational Sciences at the University of Tsukuba, Japan will have a joint meeting at the Lab this Thursday and Friday, February 24–25, 2011. The meeting will explore joint collaborations in research areas that may be of interest to both institutions, and all Computing Sciences staff are invited to attend. Here is the schedule:

Thursday, February 24, 2011, Conference Room 50B-4205

Start Time

End Time

Speaker

Affiliation

Title

8:45 

9:00 

 

 

 Opening Remarks

9:00 

9:30 

 Horst Simon

 LBNL

 Overview of Berkeley Lab

9:30 

10:00 

 Kathy Yelick

 LBNL

 Overview of Computing Sciences

10:00 

10:40 

 Mitsuhisa Sato

 Univ. of Tsukuba

 Overview of CCS

10:40 

11:00 

 

 

 BREAK

11:00 

11:40 

 Taisuke Boku

 Univ. of Tsukuba

 HA-PACS Project:
  Next Step for Scientific Frontier in CCS

11:40 

12:10 

 Hemant Shukla

 LBNL

 ICCS Overview

12:10 

13:30 

 

 

 LUNCH

13:30 

14:10 

 Tetsuya  Sakurai

 Univ. of Tsukuba

 A Scalable Parallel Method for Large-Scale Eigenvalue  Problems

14:10 

14:40 

 Jim Demmel

 UC Berkeley/  LBNL

 Optimizing Performance by Avoiding Communication

14:40 

15:10 

 Sherry Li

 LBNL

 Factorization-Based Sparse Solvers and  Preconditioners

15:10 

15:30 

 

 

 BREAK

15:30 

16:00 

 Tony  Drummond

 LBNL

 An Overview of the DOE ACTS Collection

16:00 

17:00 

 

 

 DISCUSSIONS

Friday, February 25, 2011, Conference Room 50B-4205

Start Time

End Time

Speaker

Affiliation

Title

8:45 

9:00 

 

 

 Opening Remarks

9:00 

9:40 

 Masayuki  Umemura

 Univ. of  Tsukuba

 Computational Astrophysics with a Hybrid Simulator “FIRST”

9:40 

10:10 

 Ann Almgren

 LBNL

 CASTRO: A Massively Parallel Compressible Astrophysics  Code

10:10 

10:40 

 Chao Yang

 LBNL

 Computational Techniques for Large-Scale Nuclear Structure  Calculations

10:40 

11:00 

 

 

 BREAK

11:00 

11:40 

 Kazuhiro  Yabana

 Univ. of  Tsukuba

 Quantum Dynamics Simulation in Real Time and Real Space

11:40 

12:10 

 Andrew  Canning

 LBNL

 First-Principles Materials Science Codes on Many Core  Architectures: New Algorithms and Applications

12:10 

13:30 

 

 

 LUNCH

13:30 

14:00 

 Dan Martin

 LBNL

 BISICLES: Developing a High-Performance Adaptive Ice Sheet  Model

14:00 

14:30 

 Costin Iancu

 LBNL

 Runtime and OS Support for Parallel Scientific Applications

14:30 

15:00 

 Sam Williams

 LBNL

 Auto-Tuning Memory-Intensive Kernels for Multicore

15:00 

15:20 

 

 

 BREAK

15:20 

17:00 

 

 

 DISCUSSIONS


Proposals for 2012 LDRD Funding Now Being Accepted

Lab researchers are invited to submit proposals for Laboratory Directed Research and Development (LDRD) funding, estimated at about $21 million for the FY 2012 cycle. LDRD is one of the Lab’s principal means to seed innovative science and new research directions. Cross-disciplinary proposals for projects that contribute to the Carbon Cycle 2.0 initiative, and projects from all divisions that advance the boundaries of ultrafast photon science, are strongly encouraged. The review process will use two tracks: Divisional and/or Lab-wide proposals will be reviewed by a team of senior managers, while the Discovery proposals will be reviewed by external scientific reviewers. Deadline for submission of LDRD proposals is March 18. Read more.


Two CRD Researchers Are Named 2011 Sloan Fellows

Per-Olof Persson and Koushik Sen from Berkeley Lab’s Computational Research Division have been awarded the prestigious Sloan Research Fellowship, given annually by the Alfred P. Sloan Foundation to scientists, mathematicians, and economists who are at an early stage of their research careers. The awardees will each receive a $50,000 grant over the next two years to pursue any line of research they choose. Read more.


Workshop on Exascale Data Management, Analysis, and Visualization

The DOE OASCR 2011 Workshop on Exascale Data Management, Analysis, and Visualization is being held today and tomorrow, February 22–23, in Houston, Texas. The workshop aims to bring together notable HPC scientific application scientists and experts in the fields of scientific data analysis and visualization to chart the roadmap of research necessary to support scientific discovery at the exascale.

Two Berkeley Lab staff will be giving presentations: Hank Childs will speak on “Scalable Visualization,” and John Wu will speak on “Indexing Systems.” Other LBNL participants include Wes Bethel, Surendra Byna, Keith Jackson, Terry Ligocki, Doron Rotem, and Nick Wright.


Open Cirrus Summit Issues Call for Papers

Open Cirrus is an open cloud-computing research testbed designed to support research into the design, provisioning, and management of services at a global, multi-datacenter scale. The Open Cirrus Summit organized by Yahoo!, Intel, HP, and others, will be held June 1–3, 2011 in Moscow, Russia, hosted by the Russian Academy of Sciences. The summit organizers are soliciting papers in data-intensive computing and various cloud topics; see the call for papers. Abstracts are due by March 1 and full papers by March 8. Yahoo! will award a travel grant to the student with the best paper.


“From Extreme Innovation to Extreme Collaboration” Seminar

On Tuesday, March 22, NCSA and IBM will present a seminar titled “From Extreme Innovation to Extreme Collaboration” from 9:30 am to noon in the Sutardja Dai Hall Banatao Auditorium on the UC Berkeley campus. NCSA Director Thom Dunning will review the Blue Waters project and discuss the NSF Proposal & Award Policies & Procedures for Petascale Computing Resource Allocations (PRAC). IBM representatives will give an overview of IBM research and university programs as well as the Power7 portfolio. Refreshments and lunch will be served. Click here to register.


EFRC Scientist Lecture This Friday

Berkeley Lab’s Center for Nanoscale Control of Geologic CO2 Energy Frontier Research Center (EFRC) is hosting its next lecture from 10:30 am to noon on Friday, February 25, in the Bldg. 50 auditorium. Dr. Domenik Wolff Boenisch of the Institute of Earth Sciences, University of Iceland, will speak on “Mineral Sequestration of CO2 in Basalt: A Pre-Injection Overview of the CarbFix Project.” Here is the abstract:

Carbonate minerals provide a long-lasting, thermodynamically stable and environmentally benign carbon storage host. The degree to which mineral storage is significant and the rate at which mineralization occurs depend on the rock type and injection methods. The rates could be enhanced by injecting CO2 fully dissolved in water and/or by injection into silicate rocks rich in divalent metal cations, such as basalts and peridotites. The CarbFix project (www.carbfix.com) aims at mineral sequestration of carbon in southwest Iceland early in 2011. This seminar will give details of the upcoming injection and the conditions and geochemical work behind this endeavor.

You are welcome to attend either in person or remotely. For remote audio, call 866-740-1260, access code 4865041. For remote video, go to http://lbnl.readytalk.com and use access code 4865041.


This Week’s Computing Sciences Seminars

LAPACK Seminar: Scientific Computing Challenges in the Analysis of High-Throughput Sequencing-Based Experiments
Wednesday, February 23, 11:10 am–12:00 pm, 380 Soda Hall, UC Berkeley
Lior Pachter, UC Berkeley

I will provide an introduction to recently developed “sequence-census” experiments in molecular biology, and describe the scientific computing challenges associated with interpreting the resulting data. I’ll focus on two recently developed protocols, one for measuring RNA structure and the second for measuring RNA abundance, and will discuss the large-scale convex optimization problems they motivate, and how to solve them. The work I will discuss is joint with Adam Arkin, Sharon Aviran, Julius Lucks, Adam Roberts and Cole Trapnell.

Auto-Tuned Asynchronous Hybrid Restarted Krylov Methods for Petascale Computing
Thursday, February 24, 11:00 am–12:00 pm, 50F-1647
Serge G. Petiton, Centre National de la Recherche Scientific (CNRS), Université de Science et Technologies Lille (USTL), INRIA Saclay, France

Restarted iterative Krylov methods are today classic solutions to solve large non-Hermitian sparse linear algebra problems. Hybrid asynchronous versions (based on several asynchronous parallel executions of Krylov methods with different parameters and restarted strategies) generate new parallel and distributed methods. These methods are well adapted to large clusters of clusters and future post-petascale computers. In this talk, we survey two asynchronous hybrid Krylov methods for large sparse linear algebra computing: the MERAM to compute a few eigen-elements and the GMRES/ERAM-LS method to solve linear systems. Thus, we list and analyze different parameters or processes which may be auto-tuned at runtime, such as subspace sizes, orthogonalisation methods and parameters, or sparse compression formats. Then, we present some experimental results obtained on a cluster of multi-core and on a cluster or GPU. We conclude with research projects to integrate all these results to a smart-tuned fault-tolerant GMRES-MERAM/LS method for future exascale computers.

HPC Challenges for Reactor Physics Simulations
Friday, February 25, 11:00 am–12:00 pm, 50F-1647
Christophe Calvin, Commissariat à l'Energie Atomique (CEA), Saclay, France

The aim of this talk is to present some major HPC challenges for reactor physics simulations and how these challenges are addressed at CEA/DEN. We can consider different targets of use for high performance computing in reactor physics. Depending on the target, different levels and techniques can be used. Nevertheless, the different techniques will allow us to fall back on to higher level of simulation, like:

  • Parameterized calculations: This is the basic technique for optimization. HPC is a great opportunity to take into account more parameters and to reduce “time to market.” This allows too the use of optimization techniques, like neural networks, in order to find automatically and optimize set of parameters.
  • High resolution physics: Greater memory capacity and greater CPU power are required for more refined models in each discipline. For instance deterministic transport instead of few groups diffusion approach, …
  • A more realistic physics by using systematically real physical model instead of simplified model or pre-tabulated values. This implies not only greater CPU power, but robust and easy to use coupled system.
  • Real time simulation: This already exists, but we can imagine improving modeling in order to obtain more realistic simulators and decrease the number of assumptions.

All these improvements are needed in order to increase margin by reducing uncertainties, to optimize designs, to improve safety, to optimize operating conditions and increase physics knowledge.

All these different levels of HPC will be illustrated on different kinds of applications and parallel paradigms techniques in the frame of codes developed at CEA/DEN (APOLLO3, TRIPOLI-4, ...). Results obtained for fuel load management using genetic algorithm, domain decomposition for transport solvers, GPU acceleration for the Boltzmann equation solution are given using from few cores to massively parallel computing using more than 10,000 cores.

Bridging Photonics and Computing
Friday, February 25, 11:00 am–12:30 pm, 521 Cory Hall (Hogan Room), UC Berkeley
Mario Paniccia, Intel Fellow, Director Photonics Technology Lab, Intel Corporation

The silicon chip has been the mainstay of the electronics industry for the last 40 years and has revolutionized the way the world operates. Today a silicon chip the size of a fingernail contains over one billion transistors and has the computing power that only a decade ago would take up an entire room of servers. Silicon photonics that mainly based upon silicon on insulator (SOI) has recently attracted a great deal of attention since it offers an opportunity for low cost opto-electronic solutions for applications ranging from telecommunications down to chip-to-chip interconnects as well as possible applications in new emerging areas such as optical sensing and or biomedical applications.

Recent advances and research breakthroughs in silicon photonic device performance over the last few years have shown that silicon can be considered as a material onto which one can build future optical devices. While significant efforts are needed to improve device performance and to commercialize these technologies, progress is moving at a rapid rate. If successful, silicon may similarly come to dominate the optical communications as it has the electronics industry.

This presentation will provide an overview of silicon photonics research at Intel Corporation, describe some of the recent advances, including the recently announced demonstration of an integrated silicon photonics optical link operating at 50 Gbps and the scalability of this to >1 Tbps. In addition, the presentation will provide an overview and discuss the potential applications and future opportunities for enabling photonics in and around the PC and server platform.


Link of the Week: The Software Ecosystem

In a review of the new book The Comingled Code: Open Source and Economic Development by Josh Lerner and Mark Schankerman, The Economist reports that much conventional wisdom about open source software is wrong. The authors conducted large-scale, international surveys of software users and developers, and found that the worlds of open source and proprietary software are much more “comingled” than their respective champions would have it.

The authors view the environment of software developers and users as a complex ecosystem akin to a rainforest. It would be wrong, they say, to see the two types of software as substitutes for another or as interchangeable. And free programs are not always cheaper. Read more.



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.