Computer Science

Jason M. Cope

Argonne National Laboratory O ce: 630-***-**** Mobile: (312)613-6125Mathematics and

Computer Science Division Work: *****@***.***.gov9700South Cass Ave, Building 240

Personal: *****.****@*****.**********, IL 60439 Website: http://www.mcs.anl.gov

copej/

US Citizen

Education

Ph.D. Computer Science, University of Colorado at Boulder, August 2009

Dissertation Title: Data Management for Urgent Computing Environments

Committee: Henry M. Tufo (chair), Ken Anderson, Pete Beckman, Dan Connors, and Shivakant

Mishra

M.S. Computer Science, University of Colorado at Boulder, December 2006

B.S. Computer Engineering, Virginia Polytechnic Institute and State University, May 2003,

Cum

Laude

Research Interests

High-performance computing, parallel computing, scalable systems software, parallel file

systems,

I/O forwarding, modeling HPC systems, and system software performance analysis

Research and Professional Experience

Postdoctoral AppointeeJune 2009{present

Mathematics and Computer Science Division Argonne National Laboratory

(Robert Ross and Kamil Iskra)

One of the primary developers of the I/O Forwarding Scalability Layer (IOFSL). Focused on

de-

veloping and tuning IOFSL for massively parallel systems, including IBM Blue Gene and

Cray XT

systems. Developed network layer for IOFSL that uses the native network on Cray XT

systems

(Portals API and the SeaStar network). Scaled IOFSL on a Cray XT5 system to over 193,000

cores.

Worked with collaborators to integrate IOFSL with the Vampir application performance

analysis

toolset and scaled Vampir (integrated with IOFSL) to over 200,000 cores on a Cray XT

system (a 5x

scaling improvement for the Vampir toolset). Developing exascale storage system

simulation tools

to investigate potential design points of future exascale storage systems.

Research Staff / Joint AppointmentJanuary 2010{presentComputation Institute University of Chicago

(Robert Ross and Kamil Iskra)

Developing scalable instrumentation, tracing, and performance analysis tools for HPC

applications

and I/O software stack components as part of the IOVIS project. A paper describing this

work won

a best paper award.

Student AssistantJanuary 2008{June 2009

Computational and Information Systems National Center for Atmospheric Research

Laboratory

(Henry M. Tufo)

Maintained the Coordinated TeraGrid Software and Services (CTSS) for NCAR's Frost IBM

Blue

Gene/L TeraGrid resource. Developed Grid services to support NCAR's applications and

workflows.

Represented NCAR for the TeraGrid Software Working Group.

1

Jason M. Cope

Research and Professional Experience (cont.

)

Research AssistantJanuary 2005{June 2009Department of Computer Science University of Colorado at Boulder

(Henry M. Tufo)

Developed Grid computing and Web service tools to Grid-enable Earth science applications.

Main-

tained Linux clusters and other high-performance systems. Investigated reusable software

develop-

ment techniques for Grid services. Evaluated several file systems for high-performance

computing

environments, including the Lustre, GPFS, and PVFS2 file systems. Evaluated 10 Gigabit

network

infrastructure on a regional network and the NSF TeraGrid. A paper describing this work

won first

place in a student paper competition.

Research AideMay 2008{August 2008

Mathematics and Computer Science Division Argonne National Laboratory

(Pete Beckman)

Evaluated network provisioning strategies for urgent computing workflows using Linux

network

management tools and the National Lambda Rail's Sherpa / Dynamic VLAN Service. Contin-

ued development of a discrete-event urgent computing simulator to evaluate data movement

and

placement workloads.

Givens AssociateMay 2007{August 2007

Mathematics and Computer Science Division Argonne National Laboratory

(Pete Beckman)

Integrated additional resource managers into the Special PRiority and Urgent Computing

Envi-

ronment(SPRUCE) including Cobalt and Condor. Integrated NCAR's Frost Blue Gene/L into

SPRUCE. Prototyped quality assurance and monitoring infrastructure for SPRUCE based on

Inca2.

Student Visitor May 2004{May 2007

Scientific Computing Division National Center for Atmospheric Research

(Henry M. Tufo)

Developed Web services and workflow management tools to support a Grid-enabled

biogeochemical

modeling environment (Grid-BGC). Integrated NCAR IBM Blue Gene/L system into NSF

TeraGrid,

including deployment of Grid computing software stack and development of Globus Grid

Resource

Allocation Management (GRAM) service extensions for the Cobalt resource manager.

Represented

NCAR for the TeraGrid Software Working Group.

Student AssistantAugust 2004{December 2004

Department of Computer Science University of Colorado at Boulder

(Henry M. Tufo)

Maintained Linux clusters and other high-performance systems. Evaluated several file

systems for

high-performance computing environments, including the Lustre, GPFS, and PVFS2 file

systems.

Application DeveloperMay 2003{August 2003QSS Group Inc. US Coast Guard Operation Systems Center

Developed application security and accounting tools for the United States Coast Guard's

Vessel

Documentation System.

Application DeveloperMay 2002{August 2002QSS Group Inc. US Coast Guard Operation Systems Center

Developed application management tools for the United States Coast Guard's Vessel

Documentation

System.

2

3. IOFSL Project Update, FastOS Birds-of-a-Feather Session, IEEE Supercomputing 2009 (SC

2009),

Portland, OR, November 18, 2009.

4. SPRUCE Data Management Update, IEEE Supercomputing 2009 (SC 2009), Argonne National

Laboratory Booth Presentation, Portland, OR, November 18, 2009.

4

Jason M. Cope

Presentations (cont.)

th

5. Robust Data Placement Heuristics in Urgent Computing Environments, 18 Heterogeneity in

Com-

rd

puting Workshop (HCW 2009), held during the 23 International Parallel and Distributed

Process-

ing Symposium (IPDPS 2009), Rome, Italy, May. 2009.

6. Data Management for Urgent and Emergency Computing Environments, National Institute

for Com-

putational Science Seminar, Oak Ridge National Laboratory, April 15, 2009.

7. Data Management for Urgent and Emergency Computing Environments, Mathematics and Com-

puter Science Division Seminar, Argonne National Laboratory, April 7, 2009.

8. SPRUCE Condor Demo, IEEE Supercomputing 2008 (SC 2008), Argonne National Laboratory

Booth Presentation / Demo, Austin, TX, November 19, 2008.

Grid-BGC: A Grid-

Enabled Terrestrial Carbon Cycle Modeling System, TeraGrid 2006 Conference: Advancing

Scientific

Discovery, Indianapolis, IN, June 14, 2006.

Tutorials

th

Service-Oriented Computing for HPC Data Centers and Grids, 9 LCI International Conference

on

High-Performance Clustered Computing, Urbana, IL, April 28, 2008 (1/2 day).

Software

I/O Forwarding Scalability Layer (IOFSL) - a scalable, unified high-end computing I/O

forwardingsoftware layer. Software available at http://trac.mcs.anl.gov/projects/iofsl

Portals-based network layer for BMI - Developed for use by IOFSL on Cray XT4 and XT5

systems

using SeaStar networks. Used by IOFSL at scales over 200,000 cores on a Cray XT5 system.

Softwareavailable at http://trac.mcs.anl.gov/projects/bmi

CODES - parallel discrete-event simulator for investigating exascale storage system

design points.Project description at https://press3.mcs.anl.gov/codes/

SFSCruise - an Android-based daily timber cruising program. Developed backend storage and

unit

testing infrastructure for SFSCruise.

Grid-BGC - a Grid-enabled terrestrial biogeochemical modeling environment. Project

descriptionavailable at http://www.gridbgc.ucar.edu/

Globus GRAM module for the Cobalt resource manager. Software available at

http://trac.mcs.anl.

gov/projects/cobalt/

Various software tools for the Special PRioirty Urgent Computing Environment (SPRUCE),

includ-

ing integration tools and infrastructure for Cobalt, Globus GRAM4, and Condor. Project

descriptionavailable at http://spruce.uchicago.edu/

Professional Service

th

Many

Task Computing, January, 2010.

Student Volunteer, TeraGrid 2006 Conference: Advancing Scientific Discovery,

Indianapolis, IN,

June 12 - 15, 2006.

6

Awards and Honors

Best Paper Award, 2011 Workshop on Large-Scale System and Application Performance

(LSAP2011),

June 2011.

University of Colorado Department of Computer Science's Departmental Travel Award for

presen-

tation of paper / poster at IEEE Cluster 2008, Fall 2008.

University of Colorado Department of Computer Science's Departmental Travel Award for

presen-

tation of paper at ICCS 2008, Fall 2008.

th

First Place in Student Paper Competition, The 8 LCI International Conference on Linux

Clusters:

The HPC Revolution, forExperiences Deploying a 10 Gigabit Ethernet Computing Environment

to

Support Regional Computational Science, May 2007.

University of Colorado Department of Computer Science's Clive Fraser Baillie Memorial

Travel

Award, Fall 2006.

IEEE Scalable Computing Society Travel Award, Fall 2006.

University of Colorado Department of Computer Science's Clive Fraser Baillie Memorial

Travel

Award, Fall 2005.

Eagle Scout, Boy Scouts of America, 1998.

Skills

++

C, C, MPI, OpenMP, Perl, Python, Java, bash, autotools (autoconf and automake), Linux

system

A

administration, Linux application and system software development, LT X, Globus Toolkit

v3.2

E

and 4.0 administration and development, Web services development (WSDL, WSRF), database

administration and application development (PostgreSQL, MySQL, SQLite), Matlab, Condor,

PBS

/ Torque / Maui, Cobalt, Android

7

Contact this candidate