InTheLoop | 09.29.2014
ASCR Discovery Features Trebotich's Micron-scale CO2 Models
David Trebotich, a member of the Advanced Numerical Algorithms Group, is modelling the flow of sequestered carbon dioxide at the smallest scales ever. His work is gaining some big recognition. "The models of microscopic underground pores could help scientists evaluate ways to store carbon dioxide produced by power plants, keeping it from contributing to global climate change," explains the ASCR Discovery online magazine in a recent feature on Trebotich's models. »Read more.
Trebotich's work also won this year's cover competition for the Coalition for Academic Scientific Computation's annual report, which highlights new and noteworthy achievements of universities, labs and other member organizations. The CASC report will be distributed at SC14.
ESnet’s Inder Monga Offers Views on Future Network Architectures at NORDUnet Conference
ESnet Chief Technologist Inder Monga discussed life in the fast lane in a Sept. 23 presentation at the 28th NORDUnet Conference held in Uppsala, Sweden. NORDUnet is the Nordic Infrastructure for Research & Education and is comprised of the five Nordic national research and education networks in Denmark, Finland, Iceland, Norway and Sweden. Opening the "Innovation in Networking" session, Monga's talk was titled “Life in the ‘fast’ lane: Future network architectures." Monga summarized his talk as follows: "As new approaches like Software-Defined Networking (SDN) or Named-Data Networking (NDN) appear to challenge the current way of building and operating networks, cloud providers take the lead in network innovation and conservative telcos adopt the new three letter paradigms, this talk will explore the future network architectures from the R&E perspective. The talk will also feature some of the ESnet explorations and experimentations in each of these areas." »Download Slides (PDF | 11 MB)
Former Alvarez Fellow Part of Team Recognized for HPC First
George Pau, a former recipient of the Luis W. Alvarez Postdoctoral Fellowship in Computing Sciences, was recently honored alongside his Earth Sciences Division colleagues for contributions to the Advanced Simulation Capability for Environmental Management. This marks the first attempt “to use high-performance computing uncertainty quantification to identify key controls at a contaminated site," noted an article in LANL’s Actinide Research Quarterly. During his term as an Alvarez Fellow in the Center for Computational Science and Engineering, Pau worked on high-fidelity simulations of groundwater flow related to the underground sequestration of carbon dioxide. »Read more.
Applications Open for NERSC Exascale-Science Postdoctoral Fellowships
Applications are being accepted for up to eight postdoctoral fellowship positions at NERSC. The positions are part of a larger effort to enable new, pathbreaking science with NERSC's next generation manycore Cori supercomputer. Fellows will be working in multidisciplinary teams composed of computer, computational, and domain scientists that will transition codes to the Cori system and produce mission-relevant science that truly pushes the limits of high-end computing. Fellows will be able to work on one of 20 different projects. »Learn more and apply.
Add Your Voice to The Future of the Web!
To mark the World Wide Web Consortium's (WC3) 20th anniversary, the group is hosting a three-hour symposium—called W3C20—to discuss the future of the Web at the Santa Clara Marriott on October 29, 2014. Participants will discuss how the Web community can:
- Create a more beautiful Web to enable our creative expression;
- Extend the Web to the many devices people use to improve their lives;
- Support trusted communications, secure and private; and
- Empower all people to use and contribute to the Web, including support for diverse languages and accessibility.
Tim Berners-Lee, Inventor of the World Wide Web, created W3C in October 1994. The consortium is an international community of commercial, educational, and governmental entities—including Berkeley Lab—that develops open standards to ensure the long-term growth of the Web. NERSC’s Annette Greiner is Berkeley Lab’s representative on the W3C Advisory Committee. »Learn more and register.
NERSC Aids Decades of Fusion Energy Science
Four decades of fusion energy research at NERSC have helped the plasma physics community make significant strides towards developing a practical source of fusion energy. FES-supported research at NERSC ranges from advancing the fundamental science of magnetically confined plasmas and exploring the feasibility of the inertial confinement to increasing fundamental understanding of basic plasma science to enhance economic competitiveness and create opportunities for a broader range of science-based applications. In 2013, FES-supported fusion research accounted for more than 360 million compute hours at the center in 2013, second only to the Office of Basic Energy Sciences. »Read more.
This Week's CS Seminars
Towards Large-Scale Computational Science and Engineering with Quantifiable Uncertainty
Wednesday, Oct. 1, 3:30 – 4:30 p.m., 939 Evans Hall, UC Berkeley
Tan Bui, University of Texas, Austin
We present our recent efforts towards uncertainty quantification for large-scale computational mechanics. The talk has three parts.
In the first part, we present a reduce-then-sample approach to efficiently study the probabilistic response of turbomachinery flow due to random geometric variation of bladed disk. We first propose a model-constrained adaptive sampling approach that explores the physics of the problem under consideration to build reduced-order models. Monte Carlo simulation is then performed using the cost-effective reduced model. We demonstrate the effectiveness of our approach in predicting the work per cycle with quantifiable uncertainty.
In the second part of the talk, we consider the shape inverse problem of electromagnetic scattering. We address this large-scale inverse problems in a Bayesian inference framework. Since exploring the Bayesian posterior is intractable for high dimensional parameter space and/or expensive computational model, we propose a Hessian-informed adaptive Gaussian process response surface to approximate the posterior. The Monte Carlo simulation task, which is impossible using the original posterior, is then carried out on the approximate posterior to predict the shape and its associated uncertainty with negligible cost.
In the last part of the talk, we address the problem of solving globally large-scale seismic inversion governed by the linear elasticity equation. We discuss a mesh-independent uncertainty quantification method for full wave form seismic inversion exploiting the compactness of the misfit Hessian and low rank approximation.
The Boyz From Brazil: A Case Study of Detection, Response and Remediation in Cybersecurity
Thursday, Oct. 2, 1 – 2 p.m., 250 Sutardja Dai Hall, UC Berkeley
Jim Mellander, National Energy Research Scientific Computing Center.
No abstract is available for this seminar.
CS Exascale Seminar: Computing Beyond the End of Moore’s Law
Friday, Oct. 3, 11 a.m. – 12:30 p.m., Bldg. 50B, Room 4205
John Shalf, Lawrence Berkeley National Laboratory
Moore’s Law is an observation about the economics of silicon lithography: It states that the density of devices that can be fabricated onto a silicon chip will double every 18-24 months (or conversely, that the cost per transistor will halve in that timeframe). This law has held true for the past three decades. However, conventional practices for lithography will reach atomic scale by the middle of next decade (2024-2026). At that point, it will be feasible to create lithographically produced devices with characteristic dimensions in the 3nm–5nm range. This range corresponds to a dozen or fewer Si atoms across critical device features and will therefore be a practical limit for controlling charge in a classical sense. To go further would require engineering these devices in a regime in which quantum mechanical behavior will dominate. This scaling limit will affect not only traditional CMOS technology, but all other conventional microelectronic technologies as well.
The presentation—based on a report co-written by John Shalf and Robert Leland (Sandia)—will address the following questions (which were also covered in more detail by a related OSTP report):
- What are the limits in existing semiconductor microelectronic technology that are driving the need for change? To do so, we first define some terms and then consider device-level and system-level challenges in sequence.
- What are the alternatives for future HPC technology and how do we assess their viability? Here we consider new materials and device technology, new system architectures, and new approaches to computing based on alternatives to the standard digital computation model.
Finally we offer in our conclusion a summary of major implications of the end of Moore’s Law and a perspective on the role the federal government can play in navigating this challenge.
REMOTE ACCESS INSTRUCTIONS for AUDIO & WEB CONFERENCE
To begin AUDIO conference:
- Dial Toll-Free Number: 866-740-1260 (U.S. & Canada)
- International participants dial: Toll Number: 303-248-0285, or Toll-Free: http://www.readytalk.com/intl
- Enter 7-digit access code: 4864508, followed by “#”
- Place your phone on "Mute"
To begin WEB conference:
- Go to:http://www.readytalk.com
- Enter your access code: 4864508 under “PARTICIPANT-Join