Computer Science Project

Location:

Urbana, IL

Posted:

October 15, 2012

Contact this candidate

Resume:

Gengbin Zheng

*]*.*

Department of Computer ScienceUniversity of Illinois at Urbana-Champaign

*** *. ******* ***.

Siebel Center for Computer Science, Rm 4220, UIUC

Urbana, IL 61801

1]0.8

217-***-****(o)

217-***-****(h)

******@****.***

http://charm.cs.uiuc.edu/people/gzheng

Research Interests

Parallel computing

Parallel programming languages and paradigms

Adaptive parallel runtime systems

Dynamic load balancing

Performance prediction of parallel applications

Fault Tolerance

Molecular dynamics simulation

Education

Ph.D., Computer Science, December 2005, University of Illinois at Urbana-Champaign,

Urbana, IL ; GPA 4.0

M.S., Computer Science, September 1998, Beijing University, China (with honors) ; GPA 3.8

B.S., Computer Science, September 1995, Beijing University, China; GPA 3.6

Awards

HPC Challenge class 2 Award, SC 2011, Seattle, WA

Gordon Bell Award for special accomplishment in NAMD paper, SC 2002, Baltimore, MD.

GuangHua scholarship, Study scholarship, Beijing University, 1997

LianXiang scholarship, Beijing University, 1996

LianXiang scholarship, Beijing University, 1995

Study scholarship, Beijing University, 1993, 1994

Professional Experience

Research Scientist, Computer Science Department, University of Illinois at Urbana-

Champaign, 2010-now

Research Scientist, Center for Simulation of Advanced Rockets, University of Illinois at

Urbana-Champaign, 2008-2010

Postdoctoral Research Associate, Center for Simulation of Advanced Rockets, University of

Illinois at Urbana-Champaign, 2005-2008

Graduate Research Assistant, Computer Science Department, University of Illinois at

Urbana-Champaign, 1999-2005

Intern, IBM T.J. Watson Research Center, Yorktown Heights, NY, summer 2001 - involved in

BlueGene/L (World's fastest machine) project

Intern, Silicon Graphics, Inc (SGI), in the compiler group, Mountain View, CA, summer

2000

Teaching Assistant, Computer Science Department, University of Arizona, 1998

Research assistant, Computer Science Department, Beijing University, 1995-1998

Publications

Research Center, 2001, NY

Participated Projects

Bluw Waters Project -- Port and scale Charm++ and NAMD application on Cray supercomputer

with Gemini interconnect.

NIH -- NAMD parallel molecular dynamics simulation code

NSF Next Generation Software (NGS) -- BigSim performance prediction for petaflops scale

parallel machines

DOE -- parallel rocket simulation code developed at the Center for Simulation of Advanced

Rockets (CSAR) funded by the DOE as part of its Advanced Simulation and Computing (ASCI)

program

Research and Industrial Projects

Blue Waters Project, 07 - present

The Blue Waters project aims at delivering a Cray supercomputer capable of sustained

performance of 1 petaflop on a range of real-world science and engineering applications.

It is expected to be one of the most powerful supercomputers in the world. I have been

working on this project since the project started. My work includes using BigSim

performance simulator to predict the performance of parallel applications (including NAMD)

on future Blue Waters machine, porting Charm++ runtime system on Cray Gemini interconnect

using low level Cray uGNI communication library, and scaling NAMD molecular dynamics

simulation program on this machine.

Parallel Programming Laboratory, with Laxmikant Kale, UIUC, 1/99 - present

I am one of the lead developers of Charm++ - a parallel object-oriented language and run-

time system. My work in general involves improving the performance and productivity in

high performance computing on supercomputers and workstation clusters with Charm++ run-

time system. My projects involves most aspects in Charm++ system and its applications,

including automatic dynamic load balancing techniques to improve the scalability of

parallel applications especially those challenging applications to scale on very large

parallel machines. I also work on performance tracing and analysis tool associated with

Charm++.

My Ph.D. thesis focuses on large scale parallel simulation for predicting performance of

parallel applications on extremely large parallel machines. With the simulation

infrastructure, I explore the optimization techniques needed in automatic load balancing

to improve the parallel performance on these machines.

I have been collaborating with several external groups on applications including

molecular dynamics simulation such as NAMD and LeanMD (with IBM), climate simulation

application (ISAM), and FEM applications such as Fractography3D (crack propagation

simulation) and Rocstar (Rocket Simulation).

Center for Simulation of Advanced Rockets, UIUC, 5/05 - 2010

I joined the center as a postdoc research associate, my current research topics in the

center is to exploit Charm++ in the advanced rocket simulation application to improve its

parallel performance as well as the portability. I am one of the main developers on

designing a software integration framework for multi-physics interaction and flexible high-

level orchestration modules to ease quick prototyping of coupling schemes in the rocket

simulation.

Theoretical Biophysics Group, Beckman Institute for Advanced Science and Technology,

UIUC, 1/99 - 8/04

I was one of the main developers in parallelizing NAMD application developed in the

group. NAMD is a parallel, object-oriented molecular dynamics code designed for high-

performance simulation of large biomolecular systems. Based on Charm++ parallel objects

and the load balancing framework (which I was working on), NAMD scales to hundreds of

processors on high-end parallel platforms and tens of processors on commodity clusters

using gigabit Ethernet. Our work in NAMD won the prestigious Gordon Bell Award in SC2002

for unprecedented speedup on a 3000 processor machine with peak performance of a Teraflop.

High Performance Computing Technology, development of High Performance Fortran Compiler,

with Zhuoqun Xu, Beijing University, China, 9/95 - 7/98

HPF is a data parallel programming language. The aim of this project was to design and

implement a practical HPF compiler and runtime system. My work included compiler front-end

design and implementation, SPMD source to source code translation and implementation of

the communication runtime system. This compiler was installed on DAWN1000, one of the

fastest supercomputers built in China at that time.

Parallel Large Scale Image Processing, collaborated with Chinese Academy of Science,

China, 9/97 - 7/98

In this project, we explored techniques to solve large scale image processing problems

using HPF and the compiler we developed. We developed applications in HPF for image

processing and improved the HPF compiler to achieve high performance. In order to handle

the super large scale data, which can not be completely held in main memory, we designed

and implemented parallel I/O in the HPF compiler runtime system to perform out-of-core

execution for extremely large datasets.

Synergistic Activities

Member, Association for Computing Machinery (ACM)

Program Committee: IEEE International Parallel and Distributed Processing Symposium

(IPDPS), Shanghai, China, 2012

Technical Skills

Programming: C/C++, Fortran77, Pascal, SmallTalk, Prolog, COBOL, SML, Perl, Unix Shell,

Assembly;

Parallel Programming: Fortran90, HPF, MPI, PVM, Charm++, Pthread, openMP, CORBA;

Web Programming: Java, JavaScript, HTML, CGI, Perl, Ruby, Rails, PHP;

Protocol: TCP/IP, HTTP;

OS: UNIX(many flavors), DOS/WindowsNT;

Database: Informix(4GL, ESQL), mysql;

GUI: X Window System(Xlib/Xt/Motif);

Software Tools: Lex, Yacc, CVS, Homepage design, Visual C++ ...About this document ...

Gengbin Zheng

This document was generated using the LaTeX2HTML translator Version 2008 (1.71)

University of Leeds.

Sydney.

The command line arguments were:

latex2html -no_subdir -ascii_mode -split 0 -no_navigation resume.tex

The translation was initiated by Gengbin Zheng on 2012-03-05

Gengbin Zheng2012-03-05

Contact this candidate