Employee

Aamir.Shafi@nu.edu.pk

Department of Computer Science

:(042) 111-128-128

Ext: 406

Dr. Aamir Shafi
Associate Professor

Introduction

The main objective of my research is to develop efficient and robust tools to simplify parallel application development by providing programming languages and libraries on such emerging high-end computing systems. I have had some early success by architecting and developing the open-source MPJ Express (http://mpjexpress.org) library. The MPJ Express library started off as a research prototype but has now become a robust Java MPI library with a growing community of users from leading institutions globally.

MPJ Express: I am the main architect and developer of the MPJ Express software, which is a Java-based message passing system. Such messaging libraries are typically written in languages like C, C++, and Fortran. In earlier studies, I have evaluated and compared performance of Java messaging libraries against their C counterparts. I continue to develop and maintain the MPJ Express software, which is now used in some universities and organizations for teaching and development of parallel programs. Since the first release of the MPJ Express software, around 25,000 copies (until Dec 2017) have been downloaded with a monthly average of around 250-260. More details about MPJ Express can be seen at http://mpj-express.org.

High Performance Computing Center: I also established and served as the director of the High Performance Computing center at SEECS, NUST. The main objective of this center was to conduct research in parallel computing software on emerging computing platforms. In addition, the HPC center also provided several high-end computational facilities to support innovative science and engineering projects done within Pakistan.

Current Interests and Opportunities:  My current interests revolve around exploiting High Performance Computing (HPC) techniques in optimization of deep learning applications and frameworks including Tensorflow, Caffe, Deeplearning4j. Deep learning and its applications are currently driving innovation in the HPC community. I am currently looking for students (especially Ph.D. candidates) and collaborators to join me in this effort.

Publications:

  1. Muhammad Tariq Saeed, Jamil Ahmad, Shahzina Kanwal, Andreana N. Holowatyj, Iftikhar A. Sheikh, Rehan Zafar Paracha, Aamir Shafi, et al. Formal modeling and analysis of the hexosamine biosynthetic pathway: role of O-linked N-acetylglucosamine transferase in oncogenesis and cancer progression, PeerJ 4:e2348 https://doi.org/10.7717/peerj.2348, Sept 2016.
  2. Saleh Alrashed, Jamal Alhiyafi, Aamir Shafi, Nasro Min-Allah, An efficient schedulability condition for non-preemptive real-time systems at common scheduling points, The Journal of Supercomputing, Spring US, Volume 72, Issue 12, pp 4651–4661, December 2016.
  3. Ansar Javed, Bibrak Qamar, Mohsan Jameel, Aamir Shafi, Bryan Carpenter, Towards Scalable Java HPC with Hybrid and Native Communication Devices in MPJ Express, International Journal of Parallel Programming (IJPP), Springer US, Volume 44, Issue 6, pp 1142–1172, December 2016.
  4. Hamza Zafar, Farrukh Aftab Khan, Bryan Carpenter, Aamir Shafi, Asad Waqar Malik, MPJ Express Meets YARN: Towards Java HPC on Hadoop Systems, Procedia Computer Science (International Conference on Computational Science (ICCS)), Volume 51, 2015, Pages 2678-2682, ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2015.05.379.
  5. Aamir Shafi, Aleem Akhtar, Ansar Javed, and Bryan Carpenter. 2014. Teaching parallel programming using Java.  In Proceedings of the Workshop on Education for High-Performance Computing (EduHPC 14). IEEE Press, Piscataway, NJ, USA,  56-63. DOI=10.1109/EduHPC.2014.7 http://dx.doi.org/10.1109/EduHPC.2014.7
  6. Bibrak Qamar, Ansar Javed, Mohsan Jameel, and Aamir Shafi, Design and Implementation of Hybrid and Native Communication Devices for Java HPC, Procedia Computer Science (International Conference on Computational Science (ICCS)), Volume 29, 2014, Pages 184-197, ISSN 1877-0509.
  7. Omar Khan, Mohsan Jameel, and Aamir Shafi, High Performance Message-passing InfiniBand Communication Device for Java HPC, Procedia Computer Science (International Conference on Computational Science (ICCS)), Volume 29, 2014, Pages 1468-1479, ISSN 1877-0509.
  8. Syed Abdullah Shah, Jannet Faiz, Maham Farooq, Aamir Shafi and Syed Akbar Mehdi, An Architectural Evaluation of SDN Controllers, IEEE International Conference on Communications, June 9-13, 2013, Budapest, Hungary.
  9. I-Ting Angelina Lee, Aamir Shafi, Charles E Leiserson, Memory-Mapping Support for Reducer Hyperobjects, 24th ACM Symposium on Parallelism in Algorithms and Architectures, Pittsburgh, Pennsylvania, USA, June 25 - 27, 2012. (Best Paper Award)
  10. Ammar Ahmad Awan, Muhammad Sohaib Ayub, Aamir Shafi, and Sungyoung Lee, Towards Efficient Support for Parallel I/O in Java HPC, In Proceedings of 13th International Conference on Parallel and Distributed Computing, Applications, and Technologies (PDCAT 12), Beijing, China, Dec 14th-16th, 2012.
  11. Guillermo Taboada, Juan Tourino, Ramon Doalla, Aamir Shafi, Bryan Carpenter, and Mark Baker, Device level communication libraries for high performance computing in Java, Concurrency and Computation: Practice and Experience, pp 2382-2403, 23(18), December 2011.
  12. Wakeel Ahmad, Bryan Carpenter, Aamir Shafi, Collective Asynchronous Remote Invocation (CARI): A High-Level and Efficient Communication API for Irregular Applications, Procedia Computer Science, Volume 4, 2011, Pages 26-35, ISSN 1877-0509, 10.1016/j.procs.2011.04.004.
  13. Aamir Shafi, Jawad Manzoor, Kamran Hameed, Bryan Carpenter, and Mark Baker, Multicore-enabling the MPJ Express Messaging Library, 8th International Conference on the Principles and Practice of Programming in Java (PPPJ), September 15-17 2010, Vienna, Austria
  14. Aamir Shafi, Bryan Carpenter, and Mark Baker, Nested Parallelism for Multi-core HPC Systems using Java, Journal of Parallel and Distributed Computing, pp 532-545, 69(6), June 2009, doi:10.1016/j.jpdc.2009.02.006
  15. Aamir Shafi and Jawad Manzoor, Towards Efficient Shared Memory Communications in MPJ Express, 23rd IEEE International Parallel and Distributed Processing Symposium, Rome Italy, May 2009.
  16. Aamir Shafi, Bryan Carpenter, Mark Baker, and Aftab Hussain, A Comparative Study of Java and C Performance in Two Large Scale Parallel Applications, Concurrency and Computation: Practice and Experience, http://dx.doi.org/10.1002/cpe.1416, 2009
  17. Aftab Hussain, Aamir Shafi, and Jamil Raza, Using Unified Parallel C (UPC) to Parallelize the FDTD method of Computational Electrodynamics, 13th IEEE International Conference (INMIC), pp.1-6, http://dx.doi.org/10.1109/INMIC.2009.5383096, 14-15 Dec. 2009
  18. K. Chanchio, C. Leangsuksun, H. Ong, V. Ratanasamoot, and A. Shafi, An Efficient Virtual Machine Checkpointing Mechanism for Hypervisor-based HPC System, High Availability and Performance Computing Workshop, Denver USA, 2008.
  19. Aamir Shafi, Aftab Hussain, and Jamil Raza, A Parallel Implementation of the Finite-Domain Time-Difference Algorithm using MPJ Express 22nd IEEE International Parallel and Distributed Processing Symposium, Miami Florida USA, April 2008.
  20. Mark Baker, Bryan Carpenter, and Aamir Shafi, A Buffering Layer to Support Derived Types and Proprietary Networks for Java HPC, Scalable Computing: Practice and Experience, pp 343-358, 8(4), ISSN 1895-1767, 2007.
  21. Mark Baker, Bryan Carpenter, and Aamir Shafi, MPJ Express Meets Gadget: Towards a Java Code for Cosmological Simulations, In Proceedings of the 13th European PVM/MPI Users Group Meeting (EuroPVM/MPI 2006), Bonn, Germany, September 17-20, 2006.
  22. Mark Baker, Bryan Carpenter, and Aamir Shafi, MPJ Express: Towards Thread Safe Java HPC, In Proceeding of the IEEE International Conference on Cluster Computing (Cluster 2006), Barcelona, Spain, September 25-28, 2006.
  23. Mark Baker, Matthew Grove, and Aamir Shafi, Parallel and Distributed Computing with Java, In Proceedings of the 5th International Symposium on Parallel and Distributed Computing (ISPDC 06), Timisoara, Romania, CS Press, July 6-9, 2006.
  24. Mark Baker, Bryan Carpenter, and Aamir Shafi, An Approach to Buffer Management in Java HPC Messaging, V. Alexandrov, D. van Albada, P. Sloot, and J. Dongarra (Eds), In Proceedings of the International Conference on Computational Science (ICCS 2006), LNCS. Springer, May 28-31, 2006.
  25. Aamir Shafi, Maria Riaz, Anjum Shehzad, Umer Farooq, Arshad Ali, Iosif Legrand, Harvey Newman, Distributed Agents for Mobile and Dynamics Services, In Proceedings of Computing in High Energy Physics 2003 (CHEP03), San Diego, La Jolla, California, USA, March 23-28, 2003.

Education

  • Post-doctorate(Multicore Computing), Massachusetts Institute of Technology (MIT), United States (2011)
  • PhD(Parallel Computing, High Performance Computing), University of Portsmouth, UK (2006)
  • BE(Software Engineering), National University of Sciences and Technology (NUST), Pakistan (2003)