Current Work

I am interested in:
  • Multilevel parallelism, especially hybrid programming MPI+OpenMP
    • Detecting communication and multithreading efficiency
    • Performance modeling of hybrid model
    • Interoperability of different programming models
  • Improving OpenMP scalability
    • False sharing
    • Cache reuse
  • (Semi)-Automatic OpenMP generation with the help of Compiler analysis
    • Based on analysis from OpenUH compiler: data dependence, array access pattern, cache cost model, loop cost, ...

Framework for Analyzing, Modeling and Optimizing MPI and/or OpenMP Applications

Framework for analyzing MPI/OpenMP We have developed an infrastructure for analyzing, modeling and optimizing large scale parallel applications. The framework consists of four main parts: compiler, microbenchmarks, user-interface and runtime library. Compiler generates \emph{application signature} containing loop information, parallelization status, message-passing communication and estimated execution time. Microbenchmark is needed to capture \emph{system profile} including communication latency and parallelization overhead. The user interface, based on Eclipse, is used to drive code transformation such as semi-automatic parallelization. And lastly, our runtime library can be used to balance program workload thus reducing load imbalance.

Performance Modeling for MPI and/or OpenMP Applications

Performance evaluation and modeling is a crucial process to enable the optimization of parallel programs. Programs written using two programming models, such as MPI and OpenMP, require an analysis to determine both performance efficiency and the most suitable numbers of processes and threads for their execution on a given platform. To study both of these problems, we propose the construction of a model that is based upon a small number of parameters, but is able to capture the complexity of the runtime system. We must incorporate measurements of overheads introduced by each of the programming models, and thus need to model both the network and computational aspects of the system.

We have combined two different techniques: static analysis, driven by the OpenUH compiler, to retrieve application signatures and a parallelization overhead measurement microbenchmarks to collect system profiles. Finally, we propose a performance evaluation measurement to identify communication and computation efficiency.

Publications

Journals

  1. Laksono Adhianto, Barbara M. Chapman Performance Modeling of Communication and Computation in Hybrid MPI and OpenMP Applications Simulation Modelling Practice and Theory, Vol. 15, issue 4, April 2007
  2. L.Adhianto, B. Chapman, D. Lancaster, I.Wolton, A. Kneer and F.Bodin Tools for OpenMP Application Development: The POST Project Concurrency: Practice and Experience Vol.12, pp. 1177-1191, 2000

Conferences/Workshops

  1. Laksono Adhianto, Barbara M. Chapman Autoscoping Support for OpenMP Compiler Tools and Compilers for Hardware Acceleration (TCHA'06), August 2006
  2. Laksono Adhianto, Barbara M. Chapman Performance Modeling of Communication and Computation in Hybrid MPI and OpenMP Applications In the Proceedings of the 2nd Performance Modeling and Analysis Computing (PMAC), 2006
  3. Barbara Chapman, Oscar Hernandez, Lei Huang, Tien-hsiung Weng, Zhenying Liu, Laksono Adhianto, Yi Wen Dragon: An Open64-Based Interactive Program Analysis Tool for Large Applications In the Proceedings of the 4th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT), 2003
  4. Barbara M. Chapman, T. H. Weng, Oscar Hernandez, Zhenying Liu, Yi Wen, Laksono Adhianto Cougar: Interactive Tool for Cluster Computing Proceedings of the 6th World Multiconference on Systemics, Cybernetics and Informatics, Orlando, Florida, 2002
  5. Michael Leuschel, Laksono Adhianto, Michael Butler, Carla Ferreira and Leonid Mikhailov Animation and Model Checking of CSP and B using Prolog Technology Proceedings of the ACM Sigplan Workshop on Verification and Computational Logic VCL'2001, p.97-109
  6. Laksono Adhianto and Michael Leuschel Strategy for Improving Memory Locality Reuse and Exploiting Hidden Parallelism Proceedings ISSM 2001, Manchester, August 2001
  7. M. Leuschel, I. Wolton, T. Massart, and L. Adhianto Temporal Logic Model Checking of CSP, Tools and Techniques Workshop on Automated Verification of Critical Systems, Oxford, 2001
  8. L.Adhianto, B. Chapman, D. Lancaster, I.Wolton, A. Kneer and F.Bodin Tools for OpenMP Application Development: The POST Project, First European Workshop on OpenMP (EWOMP), 1999.