Petascale Computing Resource Allocations PRAC – NSF 14-518 Ed Walker, NSF CISE/ACI March 3, 2016 1.

Slides:



Advertisements
Similar presentations
Writing Your CAREER Proposal 2014 NSF CAREER Proposal Writing Workshop University of Maryland April 7-8, 2014 CAREER Proposal Writing Workshop.
Advertisements

Extreme Scalability RAT Report title=RATs#Extreme_Scalability Sergiu Sanielevici,
Anticipated Grant Opportunities to Support Additional Time for Learning Grant Information Webinar March 14, :00 AM – 11:00 AM 2:00 PM – 3:00 PM.
1 Performance Assessment An NSF Perspective MJ Suiter Budget, Finance and Awards NSF.
User Support Coordination Objectives for Plan Years 4 and 5 (8/1/2008 to 3/1/2010) Sergiu Sanielevici, GIG Area Director for User Support Coordination.
NSF Regional Grants Conference St. Louis, MO
Tuition Reimbursement Cap Waiver Proposal MIT EMBA Class of 2014 Strategic Development & Operations Engineering June 2013.
SAN DIEGO SUPERCOMPUTER CENTER Niches, Long Tails, and Condos Effectively Supporting Modest-Scale HPC Users 21st High Performance Computing Symposia (HPC'13)
National Center for Atmospheric Research John Clyne 4/27/11 4/26/20111.
GPU System Architecture Alan Gray EPCC The University of Edinburgh.
Business & Operations Advisory Committee -- March 31, Performance Assessment.
SAN DIEGO SUPERCOMPUTER CENTER at the UNIVERSITY OF CALIFORNIA; SAN DIEGO SDSC RP Update October 21, 2010.
Merit Review and Proposal Preparation Mark Courtney Division of Environmental Biology
NSF Merit Review and Proposal Preparation Mark Courtney, Ph.D Adjunct, Department of Biology New Mexico State University 24 September 2008.
Cyberinfrastructure for Scalable and High Performance Geospatial Computation Xuan Shi Graduate assistants supported by the CyberGIS grant Fei Ye (2011)
1 Intellectual Architecture Leverage existing domain research strengths and organize around multidisciplinary challenges Institute for Computational Research.
Workshop NSF Major Research Instrumentation grants program NSF approach to research in undergraduate institutions Supporting students on grants Introduction.
National Science Foundation: Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics (TUES)
Plans for Exploitation of the ORNL Titan Machine Richard P. Mount ATLAS Distributed Computing Technical Interchange Meeting May 17, 2013.
Broader Impacts in Proposal Writing Sally Bond Assistant Director of Research Development Services Proposal Coordination Office of the Vice President for.
NSF Vision and Strategy for Advanced Computational Infrastructure Vision: NSF Leadership in creating and deploying a comprehensive portfolio…to facilitate.
1 Building National Cyberinfrastructure Alan Blatecky Office of Cyberinfrastructure EPSCoR Meeting May 21,
Partnerships and Broadening Participation Dr. Nathaniel G. Pitts Director, Office of Integrative Activities May 18, 2004 Center.
Research Support Services Research Support Services.
Extreme-scale computing systems – High performance computing systems Current No. 1 supercomputer Tianhe-2 at petaflops Pushing toward exa-scale computing.
On Preparing Proposals: Comments from Both Inside and Outside NSF Xiaodong Zhang The Ohio State University.
1 Preparing Your Application for TeraGrid Beyond 2010 TG09 Tutorial June 22, 2009.
Open Science Grid For CI-Days Elizabeth City State University Jan-2008 John McGee – OSG Engagement Manager Manager, Cyberinfrastructure.
Results of the HPC in Europe Taskforce (HET) e-IRG Workshop Kimmo Koski CSC – The Finnish IT Center for Science April 19 th, 2007.
1 Designing Effective Programs: –Introduction to Program Design Steps –Organizational Strategic Planning –Approaches and Models –Evaluation, scheduling,
ATTRACT – From Open Science to Open Innovation Information Sharing Meeting Brussels, June 19, 2014 Markus Nordberg (CERN) Development and Innovation Unit.
2009/4/21 Third French-Japanese PAAP Workshop 1 A Volumetric 3-D FFT on Clusters of Multi-Core Processors Daisuke Takahashi University of Tsukuba, Japan.
Template This is a template to help, not constrain, you. Modify as appropriate. Move bullet points to additional slides as needed. Don’t cram onto a single.
DR K-12 Program PRESENTATION HBCU-UP LEADERSHIP DEVELOPMENT INSTITUTE SPONSORED BY QEM Dr. Julia V. Clark Program Director August 13, 2009.
Joint Meeting of the AUS, US, XS Working Groups TG10 Tuesday August 3, hrs Elwood II.
NSF Cybersecurity Funding Opportunities Anita Nikolich National Science Foundation Program Director, Advanced Cyberinfrastructure September 2015.
1 NSF/TeraGrid Science Advisory Board Meeting July 19-20, San Diego, CA Brief TeraGrid Overview and Expectations of Science Advisory Board John Towns TeraGrid.
Barriers to Industry HPC Use or “Blue Collar” HPC as a Solution Presented by Stan Ahalt OSC Executive Director Presented to HPC Users Conference July 13,
Funding: Staffing for Research Computing What staffing models does your institution use for research computing? How does your institution pay for the staffing.
1Mobile Computing Systems © 2001 Carnegie Mellon University Writing a Successful NSF Proposal November 4, 2003 Website: nsf.gov.
Revision - 01 Intel Confidential Page 1 Intel HPC Update Norfolk, VA April 2008.
Arlington, VA March 31, 2004 Presentation for the Advisory Committee for Business & Operations Effective Practices Research Overview For Merit Review This.
Template This is a template to help, not constrain, you. Modify as appropriate. Move bullet points to additional slides as needed. Don’t cram onto a single.
Comprehensive Scientific Support Of Large Scale Parallel Computation David Skinner, NERSC.
National Center for Supercomputing Applications University of Illinois at Urbana–Champaign Visualization Support for XSEDE and Blue Waters DOE Graphics.
Objectives: Define and implement a website layout that enables users to find key information efficiently Develop a process to ensure the sustainability.
Course, Curriculum, and Laboratory Improvement (CCLI) Transforming Undergraduate Education in Science, Technology, Engineering and Mathematics PROGRAM.
ISTeC Research Computing Open Forum: Using NSF or National Laboratory Resources for High Performance Computing Bhavesh Khemka.
NSF Middleware Initiative Purpose To design, develop, deploy and support a set of reusable, expandable set of middleware functions and services that benefit.
Open Science Grid as XSEDE Service Provider Open Science Grid as XSEDE Service Provider December 4, 2011 Chander Sehgal OSG User Support.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views.
Tackling I/O Issues 1 David Race 16 March 2010.
CERN VISIONS LEP  web LHC  grid-cloud HL-LHC/FCC  ?? Proposal: von-Neumann  NON-Neumann Table 1: Nick Tredennick’s Paradigm Classification Scheme Early.
Scheduling a 100,000 Core Supercomputer for Maximum Utilization and Capability September 2010 Phil Andrews Patricia Kovatch Victor Hazlewood Troy Baer.
Data Infrastructure Building Blocks (DIBBS) NSF Solicitation Webinar -- March 3, 2016 Amy Walton, Program Director Advanced Cyberinfrastructure.
PEER 2003 Meeting 03/08/031 Interdisciplinary Framework Major focus areas Structural Representation Fault Systems Earthquake Source Physics Ground Motions.
General Introduction Markus Nordberg (CERN DG-DI-DI) 1
Building PetaScale Applications and Tools on the TeraGrid Workshop December 11-12, 2007 Scott Lathrop and Sergiu Sanielevici.
Benefits. CAAR Project Phases Each of the CAAR projects will consist of a: 1.three-year Application Readiness phase ( ) in which the code refactoring.
Introduction to Data Analysis with R on HPC Texas Advanced Computing Center Feb
INTRODUCTION TO XSEDE. INTRODUCTION  Extreme Science and Engineering Discovery Environment (XSEDE)  “most advanced, powerful, and robust collection.
2017 Community Grants “How to Apply” Webinar.
A Brief Introduction to NERSC Resources and Allocations
Next Step & Time plan Marcello Losasso, CERN
for the Offline and Computing groups
Cognitus: A Science Case for HPC in the Nordic Region
Tuition Reimbursement Cap Waiver Proposal
Intel® PCC Proposal Presentation
SC12 Overview \.
S-STEM (NSF ) NSF Scholarships for Science, Technology, Engineering, & Mathematics Information Materials 6 Welcome! This is the seventh in a series.
Presentation transcript:

Petascale Computing Resource Allocations PRAC – NSF Ed Walker, NSF CISE/ACI March 3,

Agenda Program description FAQs Questions and answers 2

Highest Sustained Performance: Blue Waters machine PRAC Proposals to NSF National HPC systems : XSEDE Virtual Org. XRAC proposals to xsede.org Campus machines <= focus of this presentation 3

FY 16 Solicitation (NSF ) Request computing allocations on Blue Waters You must demonstrate that – You have a breakthrough science or engineering research problem that requires petascale computing capabilities – Your code requires and can effectively exploit the capabilities offered by Blue Waters. Proposals from or including junior researchers are encouraged Next PRAC deadline is April 4,

Type of Award Compute allocation: Up to 2 years Represents significant investment of NSF  rigorous, competitive review of both the science and the suitability for execution on Blue Waters Consulting expertise from the Blue Waters team Up to $15,000 per year for travel by your team or the Blue Waters team awards anticipated Example of Blue Water projects: 5

Petascale system – sustained performance above 1 petaflop/s Deployed at the University of Illinois Tackle much larger and more complex research challenges than in the past, across a wide spectrum of domains Achieve significant impacts by – creating new knowledge about the natural world – increasing industrial competitiveness – improving national security Operational end date: expected early to mid 2018 The Blue Waters Machine 6 Images courtesy of NCSA at the University of Illinois

Blue Waters Architecture Designed for sustained petascale performance on several 100,000 cores for a broad class of problems 22,640 XE6 compute 16 AMD cores = 362,240 cores 4,224 XK compute 8 cores + 1 NVIDIA Kepler GPU PF peak performance 1.66 PB memory (Largest memory capacity of any HPC system!) I/O system: >1 TB/s bandwidth, 26.4 PB High-speed interconnect: 3D Torus, Cray Gemini Supports visualization of large datasets produced by computations More info: summaryhttp://bluewaters.ncsa.illinois.edu/hardware- summary 7

Development Plan Several years of preparation may be needed to be ready to exploit a sustained petaflop system. Present a development plan, such as: – Code scalability today – Year 1: Optimizations for scalability to part of machine – Year 2: Further optimizations for full scalability The Blue Waters team will provide consultation and support. Indicate in the plan tasks for which you need help 8

Project Description Target research problem Intellectual merit Broader impact Description of computational code(s) Development plan and source of funding Resources required – Number and type of compute nodes – Memory usage – Number and duration of runs for each development phase – Total number of node hours, I/O requirements Disclose resources allocated or proposed from other sources 9

Review Criteria Intellectual Merit and Broader Impact Is the proposed computational research leading to breakthrough science and engineering? Does it require the capability of a resource such as Blue Waters? 10

FAQ 1 Q: How can I tell if my proposed work is a good fit for the PRAC program? A: A successful PRAC proposal shows that – You have a very important science or engineering research problem that requires petascale computing capabilities – You have a solid plan to effectively exploit the massive capabilities offered by the Blue Waters machine 11

FAQ 2 Q: If PRAC does not fund the development of the computational application, who does? A: You will need to write a proposal to an NSF directorate or another agency that sponsors research in your science domain. Once you get it funded, consider writing a PRAC proposal. 12

FAQ 3 Q: Would it be better for me to write an XRAC proposal and send it to xsede.org? A: Getting an XSEDE allocation is better if – You are only beginning to parallelize your code. Request a startup allocation – You don’t have funding for developing your application into an efficient parallel code – Your code scales well, but does not need the massive capabilities of Blue Waters 13

FAQ 4 Q: What are the highlights in the recent revision of the solicitation? A: The new version revises the previous PRAC in the following ways: – All current/pending compute allocations MUST be disclosed in supplemental documentation – Total requested allocation must be stated in Project Summary – Restricts allocation request to a maximum of TWO years 14

Petascale Computing Resource Allocations (PRAC) Contact information: Ed Walker