CyberInfrastructureand

Office of CyberInfrastructure(OCI)

to: SURA Information Technology Committee Meeting

José Muñoz , Ph.D.Office of CyberInfrastructure

Deputy Office DirectorSenior Science Advisor

Muñoz 2OCI

Outline NSF organizational changes CyberInfrastructure (CI) at NSF Strategic Planning New HPC Acquisition Related OCI activities Summary

Muñoz 3OCI

CyberInfrastructure (CI) CyberInfrastructure (CI) GovernanceGovernance

• Dr. Dan Atkins (University of Michigan) selected as Dr. Dan Atkins (University of Michigan) selected as new OCI Directornew OCI Director

• served as Chair of NSF's Blue-Ribbon Advisory Panel on served as Chair of NSF's Blue-Ribbon Advisory Panel on Cyberinfrastructure: Cyberinfrastructure: Revolutionizing Science and Engineering Revolutionizing Science and Engineering Through CyberinfrastructureThrough Cyberinfrastructure (2003) (2003)• tenure begins June 2006tenure begins June 2006

• CyberInfrastructure Council (CIC) createdCyberInfrastructure Council (CIC) created• NSF ADs and ODs, chaired by Dr. Bement (NSF Dir.)NSF ADs and ODs, chaired by Dr. Bement (NSF Dir.)• CIC responsible for shared stewardship and ownership of CIC responsible for shared stewardship and ownership of NSF’s cyberinfrastructure portfolio NSF’s cyberinfrastructure portfolio

• Advisory Committee for NSF’s CI activities and Advisory Committee for NSF’s CI activities and portfolio createdportfolio created

• candidate members have been identified.candidate members have been identified.• first meeting June 2006?first meeting June 2006?

Muñoz 4OCI

CI Governance• NSF High-End Computing Coordinating GroupNSF High-End Computing Coordinating Group

• representatives from each NSF Directoraterepresentatives from each NSF Directorate• Chaired by OCIChaired by OCI• Meets every other weekMeets every other week

• SCI to OCI RealignmentSCI to OCI Realignment• SCI was moved from CISE to Office of the Director and became SCI was moved from CISE to Office of the Director and became

Office of CyberInfrastructure (OCI)Office of CyberInfrastructure (OCI)• Budget transferredBudget transferred• Ongoing projects and personnel transferredOngoing projects and personnel transferred

• OCI is focused on provision of “production-quality” CI for OCI is focused on provision of “production-quality” CI for research and educationresearch and education• CISE remains focused on CISE remains focused on basic CS research and educationbasic CS research and education

missionmission• CI software CI software andand hardware hardware• Collaborate with other NSF DirectoratesCollaborate with other NSF Directorates

• CI Strategic Planning Process UnderwayCI Strategic Planning Process Underway• NSF CyberInfrastructure Vision documentNSF CyberInfrastructure Vision document

Muñoz 5OCI

Guy Almes Program Director

Office of CyberInfrastructure

Debra CrawfordOffice Director

(Interim)

José MuñozDep. Office Dir.

Fillia MakedonProgram Director

Doug GatchellProgram Director

Kevin ThompsonProgram Director

Miriam HellerProgram Director

(Vacancy)Program Director

(Software)

Judy Hayden

Priscilla BezdekMary Daley

Irene Lombardo Allison Smith

Steve Meacham Program Director

Frank Scioli Program Director

ANL RPIU RPPU RPORNL RPTACC RPMRIREU Sites

STINMI Dev.CyberSecurity

CI-TEAMEPSCORGriPhyNDisunCCG NMI

SDSC CoreSDSC RP

HPC Acq.NCSA CoreNCSA RPPSC RP

ETF GIGEINIRNCCondorNMI Integ.Optiputer

SBE CyberToolsSBE POC Vittal Rao

Program Director

Muñoz 6OCI

CI Vision Document4 Interrelated Plans

Learning &Workforce

Development

Data, Data Analysis &

Visualization

High PerformanceComputing

Collaboratories, Observatories &Virtual Organizations

Muñoz 7OCI

Strategic Plan(FY 2006 – 2010)

• Ch. 1: Call to Action

• Ch. 2: High Performance Computing

•Ch. 3: Data, Data Analysis & Visualization

• Ch. 5: Learning and Workforce Development

• Ch. 4: Colaboratories, Observatories and Virtual Organizations

Strategic Plans for:

Completed in Summer 2006

Muñoz 8OCI

1%1% 5%1%

5%

4%

34%

14%

8%

26%

1%

CyberInfrastructure Budgets

HPC hardware acquisitions, O&M, and user support as a fraction of NSF’s overall CI budget

NSF 2006 CI Budget

75%

25%

Researchdirectorates

OCI

OCI Budget: $127M (FY06)

FY07: $182.42 (Request)

HPC ACQ.

CORE

ETF

CI-TEAM

TESTBEDS

NMI

IRNC

Muñoz 9OCI

Recent CI Activities HPC Solicitation Released September 27, 2005

Performance Benchmarks Identified November 9Proposals were due 10Feb06

Continuing Work on CI Vision DocumentFour, NSF-wide SWOT Teams developing Strategic and

Implementation Plans covering various aspects of CIhttp://www.nsf.gov/dir/index.jsp?org=OCI

Ongoing Interagency DiscussionsCommittee on ScienceOffice of Science and Technology PolicyAgency-to-agency (DARPA, HPCMOD, DOE/OS, NNSA, NIH)

Muñoz 10OCI

Acquisition Strategy

FY06 FY10FY09FY08FY07

Sc

ien

ce a

nd

en

gin

eerin

g

cap

ab

ility

(log

rithm

ic s

cale)

Typical university HPC systems

Track 1 system(s)

Track 2 systems

Muñoz 11OCI

HPC Acquisition Activities

HPC acquisition will be driven by the needs of the S&E community

RFI held for interested Resource Providers and HPC vendors on 9 Sep 2005

First in a series of HPC S&E requirements workshops held 20-21 Sep 2005Generated Application Benchmark Questionnaire Attended by 77 scientists and engineers

Muñoz 12OCI

TeraGrid: What is It?

TeraGrid: (1) Provides a unified user environment to

support high-capability, production-quality cyberinfrastructure services for science and engineering research.

(2) Provides new S&E opportunities – by making new ways of using distributed resources and services

Examples of services include:• HPC• Data collections• Visualization servers• Portals

• Integration of services provided by grid technologies

• Distributed, open architecture.• GIG responsible for integration:

• Software integration (including the common software stack, CTSS)• Base infrastructure (security, networking, and operations)• User support• Community engagement (including the Science Gateways activities)

• 8 Resource Providers (with separate awards):

• PSC, TACC, NCSA, SDSC, ORNL, Indiana, Purdue, Chicago/ANL• Several other institutions participate in TeraGrid as a sub-awardees of the GIG

• New sites may join as Resource Partners

Muñoz 13OCI

NSF Middleware Initiative (NMI)

Program Solicitations between 2001-2004 funded over 40 development awards and a series of integration awards Integration award highlights include NMI Grids Center (e.g.

Build and Test), Campus Middleware Services (e.g. Shibolleth) and Nanohub

OCI made a major award in middleware in November 2005 to Foster/Kesselman: "Community Driven Improvement of Globus Software",

$13.3M award over 5 years

Develop, deploy and sustain a set of reusable and expandable middleware functions that benefit many science and engineering applications in a networked environment. o "open standards“o international collaborationo sustainableo scalable and securable

Muñoz 14OCI

2005 IRNC Awards

AwardsTransPAC2 (U.S. – Japan and beyond)GLORIAD (U.S. – China – Russia – Korea)Translight/PacificWave (U.S. – Australia)TransLight/StarLight (U.S. – Europe)WHREN (U.S. – Latin America)

International Research Network Connections

Muñoz 15OCI

Input: 70 projects / 101 proposals / 17 (24%) projects were collaborative Input: 70 projects / 101 proposals / 17 (24%) projects were collaborative Outcomes:Outcomes:

15.7% success rate: 11 projects (14 proposals) awarded up to $250,000 over 1-2 years 15.7% success rate: 11 projects (14 proposals) awarded up to $250,000 over 1-2 years related to BIO, CISE, EHR, ENG, GEO, MPSrelated to BIO, CISE, EHR, ENG, GEO, MPS

Broadening Participation for CI Workforce DevelopmentBroadening Participation for CI Workforce Development Alvarez (FIU) – CyberBridgesAlvarez (FIU) – CyberBridges Crasta (VaTech) – Project-Centric BioinformaticsCrasta (VaTech) – Project-Centric Bioinformatics Fortson (Adler) – Fortson (Adler) – CI-Enabled 21CI-Enabled 21stst c. Astronomy Training for HS Science Teachers c. Astronomy Training for HS Science Teachers Fox (IU) - Bringing Minority Serving Institution Faculty into CI & e-Science CommunitiesFox (IU) - Bringing Minority Serving Institution Faculty into CI & e-Science Communities Gordon (OSU) – Leveraging CI to Scale-Up a Computational Science U/G CurriculumGordon (OSU) – Leveraging CI to Scale-Up a Computational Science U/G Curriculum Panoff (Shodor) – Pathways to CyberInfrastructure : CI through Computational SciencePanoff (Shodor) – Pathways to CyberInfrastructure : CI through Computational Science Takai (SUNY Stonybrook) – High School Distributed Search for Cosmic Rays Takai (SUNY Stonybrook) – High School Distributed Search for Cosmic Rays (MARIACHI)(MARIACHI)

Developing and Implementing CI Resources for CI Workforce DevelopmentDeveloping and Implementing CI Resources for CI Workforce Development DiGiano (SRI) – Cybercollaboration Between Scientists and Software DevelopersDiGiano (SRI) – Cybercollaboration Between Scientists and Software Developers Figueiredo (UFl) – In-VIGO/Condor-G MW for Coastal & Estuarine Science CI TrainingFigueiredo (UFl) – In-VIGO/Condor-G MW for Coastal & Estuarine Science CI Training Regli (Drexel) – CI for Creation and Use of Multi-Disciplinary Engineering ModelsRegli (Drexel) – CI for Creation and Use of Multi-Disciplinary Engineering Models Simpson (PSU) – CI-Based Engineering Repositories for Undergraduates (CIBER-U)Simpson (PSU) – CI-Based Engineering Repositories for Undergraduates (CIBER-U)

Learning and Our 21st Century CI WorkforceCI-TEAM: Demonstration Projects

New CI-TEAM SOLIC

ITATION COMING

New CI-TEAM SOLIC

ITATION COMING

MARCH 2006

MARCH 2006

Muñoz 16OCI

How it all fits together…

HPC

LWDCOVO

ETFCORE

CI-TEAMCI-TEAMETFIRNC

ETFCORENMI

DATA

Muñoz 17OCI

NSF HECURA 2004

FY 2004 NSF/DARPA/DOE activity focused on research in Languages Compilers Libraries

100 proposals submitted in July 2005 82 projects submitted by 57 US academic institutions and

non-profit organizations• Includes no-cost national lab and industrial lab

collaborators Nine projects were awarded

Tools and libraries for high-end computing Resource management Reliability of high-end systems

Muñoz 18OCI

NSF HECURA – 2005/2006 FOCUS

I/O, file and storage systems design for efficient, high throughput data storage, retrieval and management in the HEC environment.

hardware and software tools for design, simulation, benchmarking, performance measurement and tuning of file and storage systems.

Muñoz 19OCI

HECURA – 2005/2006 SCOPE(CISE)

File Systems Research Future File Systems related protocols I/O middleware Quality of Service Security Management, reliability, and availability at scale Archives/Backups as extensions to file systemsArchives/Backups as extensions to file systems Novel storage devices for the IO stack Novel storage devices for the IO stack I/O Architectures I/O Architectures Hardware and software tools for design, simulation of I/O, Hardware and software tools for design, simulation of I/O,

file and storage systems. file and storage systems. Efficient benchmarking, tracing, performance measurement Efficient benchmarking, tracing, performance measurement

and tuning tools of I/O, file and storage systemsand tuning tools of I/O, file and storage systems

Muñoz 20OCI

Benchmarking Broad inter-agency interest Use of benchmarking for performance prediction

valuable when target systems are not readily available either because

• Inaccessible (e.g. secure)• Does not exist at sufficient scale• In various stages of design

Useful for “what-if” analysis Suppose I double the memory on my Redstorm?

Nirvana (e.g. Snavely/SDSC): Abstract away the application: application signatures

• Platform independent Abstract away the hardware: platform signature Convolve the signatures to provide an assessment

Muñoz 21OCI

HPC Benchmarking HPC Challenge Benchmarks (http://icl.cs.utk.edu/hpcc/)

1. HPL - the Linpack TPP benchmark which measures the floating point rate of execution for solving a linear system of equations.

2. DGEMM - measures the floating point rate of execution of double precision real matrix-matrix multiplication.

3. STREAM - a simple synthetic benchmark program that measures sustainable memory bandwidth (in GB/s) and the corresponding computation rate for simple vector kernel.

4. PTRANS (parallel matrix transpose) - exercises the communications where pairs of processors communicate with each other simultaneously. It is a useful test of the total communications capacity of the network.

5. RandomAccess - measures the rate of integer random updates of memory (GUPS).

6. FFTE - measures the floating point rate of execution of double precision complex one-dimensional Discrete Fourier Transform (DFT).

7. Communication bandwidth and latency - a set of tests to measure latency and bandwidth of a number of simultaneous communication patterns; based on b_eff (effective bandwidth benchmark).

Muñoz 22OCI

DARPA HPCS Partitioned Global Address Space (PGAS)

programming paradigmIntended to support scaling to 1000s of

processorsCo-Array FortranUnified Parallel CCray’s ChapelIBM’s X10Sun’s Fortress

DARPA HPCS productivity activities HPC specific programming environments?

?

Muñoz 23OCI

OCI INVESTMENT HIGHLIGHTS

Leadership Class High-Performance Computing System Acquistion ($50M)

Data- and Collaboration-Intensive Software Services ($25.7M)Conduct applied research and developmentPerform scalability/reliability tests to explore

tool viabilityDevelop, harden and maintain software tools

and servicesProvide software interoperability

Muñoz 24OCI

HPC Acquisition - Track 1

Increased funding will support first phase of a petascale system acquisition

Over four years NSF anticipates investing $200M

Acquisition is crititical to NSF’s multi-year plan to deploy and support world-class HPC environment

Collaborating with sister agencies with a stake in HPCDARPA, HPCMOD, DOE/OS, DOE/NNSA, NIH

Muñoz 25OCI

CI – Summary“The Tide that Raises All

Boats” CI impacts and enables the broad spectrum of science and CI impacts and enables the broad spectrum of science and

engineering activitiesengineering activities NSF CI deployment/acquisition activities must be driven by

the needs of the science, engineering and education communities CI is more than just “big iron”

Many opportunities to work with other federal agencies that develop, acquire and/or use various CI resources

Work required in all aspects of CI software application and middleware for petascale systems systems software

CI has been, and will continue to be, an effective mechanism for broadening participation

Muñoz 26OCI

In the end…In the end…

It’s all about the SCIENCEIt’s all about the SCIENCE

Muñoz 27OCI