Jie Cai

Jie Cai


HPC Consultant and Developer
National Computational Infrastructure National Facility
The Australian National University
Leonard Huxley Building, Mills Road
Canberra, ACT, 0200, Australia
Phone: +61 2 6125 7965
Fax: +61 2 6125 8199
Email: email address here

Research and Work

I am interested in computer system architecture, parallel computing, cluster computing, distributed shared memory systems, as well as direct remote memory access. At August 2011, I submitted my Ph.D thesis "Region-based Techniques for Modeling and Enhancing Cluster OpenMP Performance" to ANU. The thesis was completed in the Computer System research group as a major part of Intel Cluster OpenMP project supervised by Dr. Peter Strazdins and Prof. Alistair Rendell at ANU. At December 2011, I was awarded Ph.D degree in Computer Science at ANU.

An online version of Ph.D. thesis is available at HERE!

I joined ANU Suptercomputer Facility and National Computational Infrastructure at August 2010 as a HPC Consultant and Developer. I am focusing on GPU high performance computing, parallel application (MPI and OpenMP) performance profiling and analysis, and computer system (12000+ CPUs) maintenance. In addition, I am also a co-developer of ANUPBS, a branch of OpenPBS.

I have also got research experience in sensor networks, DSP applications, and computer visualization.

Qualifications and Appointments

HPC Consultant and Developer: ANU Supercomputer Facility and National Computtational Infrastructure, (08/2010-present).

Ph.D: Research School of Computer Science, College of Engineering and Computer Science, The Australian National University, (01/2007-12/2011)

Research Student: School of Electrical and Information Engineering, University of Sydney, (07/2006-01/2007)

Master Degree: Master of Information Technology-Advanced (by research), Griffith University (02/2004-02/2006)

Undergraduate: B.E. in Automation, Department of Automation, University of Science and Technology of China (USTC) (09/1998- 07/2003)

Awards and Prizes

Teaching

Publications

Under Review

  1. Jie Cai and Peter E. Strazdins, "An Accurate Prefetch Technique for Dynamic Paging Behaviour for Software Distributed Shared Memory", under review.

Peer-reviewed Conference Papers

  1. Peter E. Strazdins, Jie Cai, Muhammad Atif, and Joesph Antony, "Scientific Application Performance on HPC, Private and Public Cloud Resources: A Case Study Using Climate, Cardiac Model Codes and the NPB Benchmark Suite", Proceedings of IPDPS, 2012.
  2. Jie Cai, Peter E. Strazdins, and Alistair P. Rendell, "Region-Based Prefetch Techniques for Software Distributed Shared Memory Systems", 10th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid 2010), page 113-122, May 2010.
  3. Jie Cai, Alistair P. Rendell, and Peter E. Strazdins, "Non-threaded and Threaded Approaches to Multirail Communication with uDAPL", 6th IFIP/IEEE International Conference on Network and Parallel Computing (NPC 2009), page 233-239, Oct. 2009.
  4. Rui Yang, Jie Cai, Alistair P. Rendell, and V Ganesh, "Use of Cluster OpenMP with the Gaussian Quantum Chemistry Code: A Preliminary Performance Analysis", 2009 International Workshop on OpenMP (IWOMP 2009), page 53-62, LNCS 5568 (2009).
  5. J. Cai, A. P. Rendell, P. Strazdins, and H.J. Wong, "Performance Models for Cluster-Enabled OpenMP Implementations", Proceeding of 13th IEEE Asia-Pacific Computer Systems Architecture Conference (ACSAC 2008), page 1-8, 4-6 Aug. 2008.
  6. H.J. Wong, J. Cai, A.P. Rendell and P. Strazdins,"Micro-Benchmarks for Cluster OpenMP Implementations: Memory Consistency Costs", 2008 International Workshop on OpenMP (IWOMP 2008), LNCS 60-70, 5004 (2008).
  7. J. Cai, D. C. Levy "Using Stationary Electronic Noses Network to Locate Dynamic Odour Source", Proceeding of 2007 IEEE International Conference On Integration Technology (IEEE ICIT07), ISBN: 1-4244-1092-4, pp793-798, 2007.
  8. R. Sitte, J. Cai, "Visualizing Dynamic Etching in MEMS VR-CAD Tool", Proceedings of The 14-th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision'2006, Plzen (Czech Rep.) ISBN 80-86943-03-8, pp 343-350. This publication was accepted as full paper. (triple, blind, full paper reviewing process, 20.9 acceptance rate), available online http://wscg.zcu.cz/wscg2006/Papers_2006/Full/!WSCG2006_Full_Proceedings_Final.pdf.

Journal Papers

  1. J. Cai, D. C. Levy, "Source Direction Detection Based on Stationary Electronic Nose System", International Journal of Applied Mathematics and Computer Sciences, Volume 3, Number 4, pp 179-184, available online: http://www.waset.org/journals/ijamcs/v3/v3-4-32.pdf, 2007. [also see: Transaction of World Academy of Science, Engineering and Technology, Volume 15, pp 115-119, ISBN: 975-00803-4-3 /ISSN 1305-5313, available online: http://www.waset.org/pwaset/v15/v15-22.pdf].
  2. J. Cai, D. C. Levy, "Tracking Dynamic Source Direction with a Novel Stationary Electronic Nose System", Sensors (ISSN: 1424-8220), issue 11, vol 6, pp 1537-1554, 2006, available online http://www.mdpi.org/sensors/list06.htm#issue11.

Peer Reviewed Posters

  1. Jie Cai, Muhammad Atif, Michael Chapman, Joseph Anthony, David Singleton, Ben Evans, "DCC: Data Centric Compute Resource at the NCI-NF", 4th IEEE/ACM International Conference on Utility and Cloud Computing (UCC 2011), December 2011.

Technical Report

  1. J. Cai, A. P. Rendell, P. Strazdins, and H.J. Wong, "Predicting Performance of Intel Cluster OpenMP with Code Analysis Method", ANU Computer Science Technical Reports, TR-CS-08-03, 2008.

Other Talks and Presentations

  1. Jie Cai, "Performance Tools @ NCI NF Systems", ANU Supercomputer Facility, Feb, 2011.
  2. Jie Cai, "GPU High Performance Computing @ ANUSF", ANU Supercomputer Facility, Sept., 2010.
  3. Jie Cai, "Cluster OpenMP System: A Middleware for both Programmability and Performance", High Research Degree Poster, SoCS, CECS, ANU, July, 2010.
  4. Jie Cai, Peter Strazdins, Alistair Rendell,"Region-Based Prefetch Techniques for Software Distributed Shared Memory Systems", 10th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid 2010), Melbourne, Australia, May 2010.
  5. Jie Cai, Alistair Rendell, Peter Strazdins, "Non-threaded and Threaded Approaches to MultiRail Communication with uDAPL", 6th IFIP Network and Parallel Computing Conference, Gold Coast, Australia, Oct. 2009.
  6. Jie Cai, "Programmability and Performance: OpenMP Implementation on Cluster Systems", High Research Degree Poster, SoCS, CECS, ANU, June, 2009.
  7. Jie Cai, "User Direct Access Programming Library (uDAPL)", Department of Computer Science, the Australian National University, December, 2008.
  8. Jie Cai, Alistair Rendell, Peter Strazdins, H'Sien Jin Wong, "Peformance Models for Cluster-Enabled OpenMP Implementations", 13th IEEE Asia-Pacific Computer Systems Architecture Conference, August, 2008.
  9. Jie Cai, "OpenMP Implementation on Advanced Cluster Systems", DCS SEMINAR SERIES, the Australian National University, May, 2008.
  10. Jie Cai, Peter Strazdins, Alistair Rendell, H'Sien Jin Wong, "Performance Evaluation for Intel Cluster OpenMP", Student paper presentation, APAC'07, Perth, Australia, 2007.

Benchmarks and Codes

  1. Peter Strazdins and Jie Cai,"Optimized LINPACK OpenMP Benchmark for CLOMP", 2010.
  2. Peter Strazdins and Jie Cai,"LINPACK OpenMP Benchmark", 2009.
  3. Jie Cai, Alistair Rendell, and Peter Strazdins, "uDAPL RDMA MultiRail Bandwidth Benchmarks", 2009.
  4. Jie Cai, "A simple FFT benchmark for sDSM", 2008.
  5. Jie Cai, "A simple Laplace benchmark for sDSM", 2008.
  6. Jie Cai, Peter Strazdins, Alistair Rendell, H'Sien Jin Wong, "SEGVCOST: WRITE and FETCH Fault Servicing Cost Benchmark for Cluster-enabled OpenMP Systems", 2008.
  7. H'sien Jin Wong, Jie Cai, Alistair Rendell, and Peter Strazdins, "MCBENCH: Memory Consistency Benchmark for Cluster-enabled OpenMP Systems", 2008.
Locations of visitors to this page
Last modified: Mon May 10 20:59 EST 2010