Fengmei Zhao
Sugar Land, TX ****9
Phone: 281-***-****
***********@*******.***
OBJECTIVE
A POSITION THAT I CAN CONTRIBUTE WELL WITH MY EDUCATION BACKGROUND AND
PROGRAMMING EXPERIENCE
in high performance computing and compiler technology.
Experience
RESEARCH PROGRAMMER, RICE UNIVERSITY; HOUSTON, TX - 2001-2010
HPC Challenge benchmarks in CoArray Fortran 2.0 - 2009-2010
Participate development and performance tuning of HPC Challenge benchmarks.
HPC Challenge benchmarks suite is a collection of kernels, each with
different sensitivities to computational throughput, memory latency, memory
bandwidth, communication latency, and communication bandwidth. CAF 2.0
implementation of HPCC benchmarks include STREAM, RandomAccess, FFT, and
High Performance Linpack (HPL).
ROSE Compiler - 2008-2010
Senior developer of the Rose compiler project at the Lawrence Livermore
National Laboratory. Participate development of Fortran front-end and
intermediate representation and un-parser. Contribution is recorded at
https://outreach.scidac.gov/scm/?group_id=24. ROSE is an open source
compiler infrastructure to build source-to-source program transformation
and analysis tools for large-scale Fortran 77/95/2003, C, C++, OpenMP, and
UPC applications.
CoArray Fortran Compiler - 2007-2010
Main developer of CoArray Fortran compiler. CoArray Fortran is a SPMD
parallel programming model based on a small set of language extensions to
Fortran 90. CoArray Fortran supports access to non-local data using a
natural extension to Fortran 90 syntax, synchronization primitives,
pointers and dynamic allocation of shared data, and parallel I/O.
HPC Performance Toolkit - 2006-2007
Participate development of HPCToolkit. HPCToolkit is a suite of multi-
platform tools for performance analysis of sequential and parallel
programs. The tools are designed for analyzing the node performance of
optimized application binaries without prior arrangement.
Open-source Compiler Technology for Source-to-Source Optimization - 2001-
2007
Main developer of Open-source compiler Open64/SL for CAF 1.0. The Open64/SL
project at Rice University is a project to adapt the Open64 compiler
infrastructure, released as open source by SGI, into an infrastructure to
support source-to-source transformation of production programs. The Open64
infrastructure includes a near commercial-quality front end for Fortran 90
from Cray and gcc-based front ends for C and C++.
Intermediate Form Independent Program Analysis - 2001-2005
Participate development of the OpenAnalysis project. The OpenAnalysis open-
source project grew out of an effort to build some simple compiler-
infrastructure independent analyses that were suitable for use with both
abstract syntax tree level intermediate representations and machine-code
level representations.
Programming Course Projects, Rice University; Houston, TX - 2002-2007
Parallel Implementation of Lindsey-Fox Algorithm - 2006
Implement a parallel program for factoring high degree polynomials in
signal processing using Matlab distributed computing toolbox based on the
Lindsey-Fox algorithm. This implementation achieved up to 11-fold speedup
on factoring real random coefficients polynomials of degree 600000 ran on
up to 24 CPUs of a Cray XD1 cluster. Experimental results were presented by
Burrus et al. in a paper: A parallel version of the Lindsey-Fox algorithm
for factoring high degree polynomials in signal processing, at the 12th
Digital Signal Processing Workshop, 2006.
Parallel Implementation of board game in Cilk - 2006
Implement a board game in parallel with Cilk.
Implementation of a unix-style OS kernel and file system - 2007
Implement a unix-style OS kernel and file system in C.
Software Implementation for phylogenetic analysis using maximum parsimony -
2006
Participate implementation of programs in C++ and MPI for phylogenetic
analysis using maximum parsimony. The work led to a published paper:
Reconstructing phylogenetic networks using maximum parsimony, in the 2005
IEEE Computational Systems Bioinformatics Conference.
Education
RICE UNIVERISTY, COMPUTER SCIENCE DEPARTMENT, HOUSTON, TEXAS, 2001 - 2008.
FINISH 13 GRADUATED LEVEL COURSES: COMPUTER SYSTEM ARCHITECTURE, MULTI-CORE
COMPUTING, PARALLEL COMPUTING, COMPUTER SYSTEM PERFORMANCE, BIOINFORMATICS,
DESIGN AND ANALYSIS OF ALGORITHMS, COMPILER CONSTRUCTION, ADVANCED COMPILER
CONSTRUCTION, OPERATING SYSTEMS AND CONCURRENT PROGRAMMING, ADVANCED
COMPILATION FOR VECTOR PARALLEL PROCESSORS, DESIGNING EMBEDDED COMPUTING
ENVIRONMENTS, AUTOMATA AND FORMAL LANGUAGE AND COMPATIBLE, LOGIC IN
COMPUTER SCIENCE. GPA 3.76/4.0.
Changsha Institute of Technology, Changsha, China - Computer Science, 1991-
1995. Finish all required classes and exams for PhD in 1993 with GPA
4.0/4.0. Drop out of the program in 1995.
Changsha Institute of Technology, Changsha, China - Computer Science, top
1% graduate student in the department with GPA 4.0/4.0, 1989 - 1991.
Nankai University, Tianjin, China - B. Sc. Computer Science and
Mathematics, GPA 3.6/4.0, 1985
Skills
PROGRAMMING LANGUAGES: C/C++, FORTRAN, JAVA, PERL, SHELL SCRIPT, ASSEMBLY,
MATLAB.
Parallel computing: MPI, Coarray Fortran, OpenMP, Cilk, UPC for multi-
threading, multi-core, distributed system, cluster programming.
OS and environments: Linux/Unix, MacOS, Windows, Gnu System.
References
LAKSONO ADHIANTO, PHD, SENIOR RESEARCH PROGRAMMER ANALYST, COMPUTER SCIENCE
DEPARTMENT, RICE UNIVERSITY, *******@****.***, 713-***-****.
Cristian Coarfa, PhD, Assistant Professor, Baylor College of Medicine,
******@***.***, 713-***-****.
Yuri Dotsenko, PhD, Senior Software Development Engineer, Microsoft,
******@*********.***, 425-***-****.
Yuan Zhao, PhD, Software Engineer, Texas Instruments, ********@**.***, 281-
274-2855.