Page 68 - 2017 Brochure
P. 68
究員
李丕榮 Pei Zong Lee
Research Fellow
Ph.D., Computer Science, New York University
Tel: +886-2-2788-3799 ext. 1812 Fax: +886-2-2782-4814
Email: leepe@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/leepe/
• Research Fellow, Institute of Information Science, Academia Sinica (1998-present)
• Associate Professor (Adjunct), Department of Electrical Engineering, National Chung Cheng University (1993-1996)
• Associate Research Fellow, Institute of Information Science, Academia Sinica (1989-1998)
• Ph.D., Computer Science, New York University (1986-1989)
• M.S., Computer and Decision Sciences, National Tsing Hua University (1982-1984)
• B.S., Mathematics, National Taiwan University (1975-1979)
Research Description subscript array of arrays or pointers to implement indirect memory
accesses for irregular computation, compilers cannot determine
My research interests are in compilers for scientific applications, either data dependence relations or data alignment relations at
parallel algorithm design, computer architectures, and the interplay the compiling time. As such, if compilers only provide naïve data
among architectures, algorithms and compilers. A major challenge distributions, generated code cannot avoid adopting expensive
with implementing large scientific applications on current parallel communication primitives, which lengthen the execution time and
computers lies in handling data locality among memory hierarchies, thus degrade the performance of parallel processing.
such that memory conflicts are avoided on shared memory parallel
computers and communication overhead is avoided on distributed To understand the techniques for compiling irregular computation,
memory parallel computers. Compilers act as bridges connecting we analyze real code for scientific computation. We conduct
algorithms and architectures, and I am interested in studying the research on 1) unstructured mesh generation 2) unstructured mesh
interdependence of these three entities. partition 3) Euler equation and Navier-Stokes equation solvers for
a numerical wind tunnel platform, an engine combustion platform
When running a program that does not use indirect memory for computing reactive flows 4) visualization and 5) the use of
accesses, such as a subscript array of arrays or pointers, on MPI on multi-core workstation/PC clusters to accelerate irregular
distributed memory parallel computers, compilers can find data computation. Our goal is to provide paradigms for designing
dependence relations among statements and data alignment different classes of computation-intensive applications, such that
relations among data arrays. Therefore, compilers have enough compilers can automatically generate efficient code for parallel
information to determine data distribution, execution scheduling, computation.
and the generation of communication code using effective
communication primitives. However, if a program does use
Publications 7. PeiZong Lee and Wen-Yao Chen. 2002. “Generating Communication
Sets of Array Assignment Statements for Block-Cyclic Distribution on
1. PeiZong Lee and Zvi M. Kedem. 1988. “Synthesizing Linear-Array Distributed Memory Parallel Computers,” Parallel Computing, Vol.
Algorithms from Nested For Loop Algorithms,” in The Special 28, No. 9, Sep. 2002, page 1329--1368.
Issue on Parallel and Distributed Algorithms, IEEE Transactions on
Computers, Vol. C-37, No. 12, Dec. 1988, page 1578--1598. 8. PeiZong Lee and Zvi M. Kedem. 2002. “Automatic Data and
Computation Decomposition on Distributed Memory Parallel
2. PeiZong Lee and Zvi M. Kedem. 1990. “Mapping Nested Loop Computers,” ACM Transactions on Programming Languages and
Algorithms into Multidimensional Systolic Arrays,” in IEEE Systems, Vol. 24, No. 1, Jan. 2002, page 1--50.
Transactions on Parallel and Distributed Systems, Vol. 1, No. 1, Jan.
1990, page 64--76. 9. PeiZong Lee, Chih-Hsueh Yang, and Jeng-Renn Yang. 2004.
“Fast Algorithms for Computing Self-Avoiding Walks and Mesh
3. PeiZong Lee and Fang-Yu Huang. 1994. “Restructured Recursive Intersections over Unstructured Meshes,” Advances in Engineering
DCT and DST Algorithms,” in IEEE Transactions on Signal Software, Vol. 35, No. 2, Feb. 2004, page 61—73.
Processing, Vol. 42, No. 7, July 1994, page 1600--1609.
10. PeiZong Lee, Chien-Min Wang, and Jan-Jan Wu. 2006. “Compiler
4. PeiZong Lee and Fang-Yu Huang. 1994. “An Efficient Prime-Factor and Run-time Parallelization Techniques for Scientific Computations
Algorithm for the Discrete Cosine Transform and Its Hardware on Distributed Memory Parallel Computers,” included in the book
Implementations,” IEEE Transactions on Signal Processing, Vol. 42, High Performance Computing: Paradigm and Infrastructure, pp. 135--
No. 8, Aug. 1994, page 1996--2005. 181, edited by Dr. Laurence T. Yang and Dr. Minyi Guo, John Wiley
& Sons, Inc., 2006.
5. PeiZong Lee. 1995. “Techniques for Compiling Programs on
Distributed Memory Multicomputers,” Parallel Computing, Vol. 21,
No. 12, Dec. 1995, page 1895--1923.
6. PeiZong Lee. 1997. “Efficient Algorithms for Data Distribution on
Distributed Memory Multicomputers,” IEEE Transactions on Parallel
and Distributed Systems, Vol. 8, No. 8, Aug. 1997, page 825--839.
66 研究人員 Research Faculty
李丕榮 Pei Zong Lee
Research Fellow
Ph.D., Computer Science, New York University
Tel: +886-2-2788-3799 ext. 1812 Fax: +886-2-2782-4814
Email: leepe@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/leepe/
• Research Fellow, Institute of Information Science, Academia Sinica (1998-present)
• Associate Professor (Adjunct), Department of Electrical Engineering, National Chung Cheng University (1993-1996)
• Associate Research Fellow, Institute of Information Science, Academia Sinica (1989-1998)
• Ph.D., Computer Science, New York University (1986-1989)
• M.S., Computer and Decision Sciences, National Tsing Hua University (1982-1984)
• B.S., Mathematics, National Taiwan University (1975-1979)
Research Description subscript array of arrays or pointers to implement indirect memory
accesses for irregular computation, compilers cannot determine
My research interests are in compilers for scientific applications, either data dependence relations or data alignment relations at
parallel algorithm design, computer architectures, and the interplay the compiling time. As such, if compilers only provide naïve data
among architectures, algorithms and compilers. A major challenge distributions, generated code cannot avoid adopting expensive
with implementing large scientific applications on current parallel communication primitives, which lengthen the execution time and
computers lies in handling data locality among memory hierarchies, thus degrade the performance of parallel processing.
such that memory conflicts are avoided on shared memory parallel
computers and communication overhead is avoided on distributed To understand the techniques for compiling irregular computation,
memory parallel computers. Compilers act as bridges connecting we analyze real code for scientific computation. We conduct
algorithms and architectures, and I am interested in studying the research on 1) unstructured mesh generation 2) unstructured mesh
interdependence of these three entities. partition 3) Euler equation and Navier-Stokes equation solvers for
a numerical wind tunnel platform, an engine combustion platform
When running a program that does not use indirect memory for computing reactive flows 4) visualization and 5) the use of
accesses, such as a subscript array of arrays or pointers, on MPI on multi-core workstation/PC clusters to accelerate irregular
distributed memory parallel computers, compilers can find data computation. Our goal is to provide paradigms for designing
dependence relations among statements and data alignment different classes of computation-intensive applications, such that
relations among data arrays. Therefore, compilers have enough compilers can automatically generate efficient code for parallel
information to determine data distribution, execution scheduling, computation.
and the generation of communication code using effective
communication primitives. However, if a program does use
Publications 7. PeiZong Lee and Wen-Yao Chen. 2002. “Generating Communication
Sets of Array Assignment Statements for Block-Cyclic Distribution on
1. PeiZong Lee and Zvi M. Kedem. 1988. “Synthesizing Linear-Array Distributed Memory Parallel Computers,” Parallel Computing, Vol.
Algorithms from Nested For Loop Algorithms,” in The Special 28, No. 9, Sep. 2002, page 1329--1368.
Issue on Parallel and Distributed Algorithms, IEEE Transactions on
Computers, Vol. C-37, No. 12, Dec. 1988, page 1578--1598. 8. PeiZong Lee and Zvi M. Kedem. 2002. “Automatic Data and
Computation Decomposition on Distributed Memory Parallel
2. PeiZong Lee and Zvi M. Kedem. 1990. “Mapping Nested Loop Computers,” ACM Transactions on Programming Languages and
Algorithms into Multidimensional Systolic Arrays,” in IEEE Systems, Vol. 24, No. 1, Jan. 2002, page 1--50.
Transactions on Parallel and Distributed Systems, Vol. 1, No. 1, Jan.
1990, page 64--76. 9. PeiZong Lee, Chih-Hsueh Yang, and Jeng-Renn Yang. 2004.
“Fast Algorithms for Computing Self-Avoiding Walks and Mesh
3. PeiZong Lee and Fang-Yu Huang. 1994. “Restructured Recursive Intersections over Unstructured Meshes,” Advances in Engineering
DCT and DST Algorithms,” in IEEE Transactions on Signal Software, Vol. 35, No. 2, Feb. 2004, page 61—73.
Processing, Vol. 42, No. 7, July 1994, page 1600--1609.
10. PeiZong Lee, Chien-Min Wang, and Jan-Jan Wu. 2006. “Compiler
4. PeiZong Lee and Fang-Yu Huang. 1994. “An Efficient Prime-Factor and Run-time Parallelization Techniques for Scientific Computations
Algorithm for the Discrete Cosine Transform and Its Hardware on Distributed Memory Parallel Computers,” included in the book
Implementations,” IEEE Transactions on Signal Processing, Vol. 42, High Performance Computing: Paradigm and Infrastructure, pp. 135--
No. 8, Aug. 1994, page 1996--2005. 181, edited by Dr. Laurence T. Yang and Dr. Minyi Guo, John Wiley
& Sons, Inc., 2006.
5. PeiZong Lee. 1995. “Techniques for Compiling Programs on
Distributed Memory Multicomputers,” Parallel Computing, Vol. 21,
No. 12, Dec. 1995, page 1895--1923.
6. PeiZong Lee. 1997. “Efficient Algorithms for Data Distribution on
Distributed Memory Multicomputers,” IEEE Transactions on Parallel
and Distributed Systems, Vol. 8, No. 8, Aug. 1997, page 825--839.
66 研究人員 Research Faculty
