PhD Student at Singapore Management University
Mobile +65 8284 2620
Email towernguyenminh [at] gmail [dot] com
Skype towernguyenminh
Research Interests
Social Network Mining
Large-scale Data Mining
High Performance Computing
Education
I'm currently PhD student at School of Information Systems, Singapore Management University.
2006 - 2011: I got my BE from Honor Program at Hanoi University of Science and Technology (HUST) one of the most prestige university in Vietnam.
GPA: 8.22/10 (Excellent Grade)
2003 - 2006: I went to High School for Gifted Students (HSGS) which belongs to Hanoi University of Science (HUS), Vietnam National University (VNU) at Hanoi.
Research Experience
August 2012 - Now: Work at Living Analytics Research Centre, School of Information Systems, Singapore Management University
July 2011 - April 2012: I worked at VNG Corporation, one of the largest Internet company in Vietnam. The project I worked on is to build a Feed Ranking System.
2009 - 2011: I worked at HPC Center, HUST where I've learned a lot of precious knowledge about Parallelism, Cluster Computing with MPI, GPU Computing with CUDA, Heterogeneous GPU+CPU Cluster Computing.
My very first experience (2009) with Parallel Computing is the project Parallel Quicksort which implements two algorithms: Hyper-cube and Sample pivot. It is interesting that the latter algorithm is very similar to GPU Merge sort by N. Satish et al. in "Designing Efficient Sorting Algorithms for Manycore GPUs".
Publications
An Intermediate Library for Multi-GPUs Computing Skeletons, IEEE International Conference on Computing and Communication Technologies (RIVF) 2012 (accepted).
Abstract: This paper introduces a library which supports programmers to write parallel programs on GPU architecture, especially with a system consisting of multi-GPUs. The library is designed based on skeletons, which helps us to make parallel programs easily and quickly as if writing sequential programs. Skeletons usually are described by functional language which supports high-order function totally. Because of the performance and popularity of C++ language, we try to re-annotate C++ language to support high-order functions completely, hence, it is convenience for us to create general-purpose skeletons.
CUSKEL – a Parallel Skeleton Library for Multi-GPU, Fundamental and Applied IT Research National Conference (FAIR) 2011, Bien Hoa, Vietnam.
Abstract: Nowadays, GPU computing is known as a significant way to achieve high performance programs. Although it is not expensive to have many GPUs to enhance the speed of a program, the problem of writing a correct and efficient program is still a great challenge to programmers. Our work focuses on building a skeletal parallel library, namely CUSKEL, which allows people to write their GPU-based parallel programs as easily as writing CPU-based sequential programs. The library can automatically choose the best configuration of GPU’s threads and blocks to obtain best performance on a specific GPU. By dint of theory of Constructive Algorithmics, our library can easily expand to a large range of skeletons from basic ones, depending on application-specific domains. The code using our library often runs tens to hundreds times faster than sequential code on our experimentation system.
Pkware Cryptabalysis Using a Cluster of Graphics Cards, Southeast Asian Conference (SEATUC) 2011, Hanoi, Vietnam.
Abstract: In this paper, we present an approach using modern multi-core graphic processing units (GPUs) as the computing device for finding lost passwords of ZIP archives. Our algorithm includes three phases: first, we generate the password search space by the exhaustive approach; then we reduce the original very large password search space to a much smaller one which surely containing the correct password. Based on reduced set of passwords we decrypt encrypted the ZIP file. These checking, decryption processes are implemented on GPUs. Finally we decompress and apply plain text recognition for the decrypted data to find out the correct password of the ZIP file, this process is implemented on CPU.
ETS Tests
Toefl: 93 (Reading 29, Listening 25, Writing 22, Speaking 17).
GRE: 1210 (Verbal 410, Quantitative 800).
Technical Skills
Programming languages: C/C++, Java, Python
Parallel Programming Framework/Interface: MPI, CUDA, MapReduce, GraphLab
References
Ee-Peng LIM, Professor
School of Information Systems, Singapore Management University
Jing JIANG, Assistant Professor
School of Information Systems, Singapore Management University
NGUYEN Huu-Duc, PhD
Director of High Performance Computing Center, Hanoi University of Science and Technology
.jpg){kind=small}