academic.info

My personal web page at CSLab

research.interests

My research interests lie mainly in the following areas:

  • High performance computing
  • Scientific applications
  • Multicore architectures and optimizations
  • Parallel programming
  • Power-aware computing
  • Machine learning

My Ph.D. thesis focused on the optimization of the Sparse Matrix-Vector Multiplication kernel (SpMV) for modern multicore architectures. I performed an in-depth performance analysis of the kernel and identified its major performance bottlenecks. This allowed our team to propose an advanced storage format for sparse matrices, the Compressed Sparse eXtended (CSX) format, which targets specifically the minimization of the memory footprint of the sparse matrix. This format provides significant improvements in the performance of the SpMV kernel in a variety of matrices and multicore architectures, maintaining a considerable performance stability. Finally, I investigated the performance of the SpMV kernel from an energy-efficiency perspective, in order to identify the execution configurations that lead to optimal performance-energy tradeoffs. A list of related publications can be found here.

In the past, I have also been interested in intelligent systems and pattern recognition algorithms. In particular, I have worked on data classification algorithms using concepts from artificial immune systems. A list of publications in this field can be found here.

You can also check my Google scholar account for statistics on my publications.

phd.thesis

  • V. Karakasis.Optimizing the Sparse Matrix-Vector Multiplication Kernel for Modern Multicore Computer Architectures. Ph.D. Thesis, School of ECE, National Technical University of Athens, 2012. [english], [greek]

publications.hpc.list

  • J. C. Meyer, V. Karakasis, J. Cebrián, L. Natvig, D. Siakavaras, and K. Nikas. Energy-efficient sparse matrix autotuning with CSX – A trade-off study. In Ninth Workshop on High-Performance, Power-Aware Computing (HPPAC'13), IPDPS'13, Boston, MA, USA, 2013. IEEE. [pdf]
  • V. Karakasis, T. Gkountouvas, K. Kourtis, G. Goumas, and N. Koziris. An extended compression format for the optimization of sparse matrix-vector multiplication. IEEE Transactions on Parallel and Distributed Systems (TPDS), 24(10):1930–1940, 2013. IEEE. [pdf], [pub]
  • T. Gkountouvas, V. Karakasis, K. Kourtis, G. Goumas, and N. Koziris. Improving the performance of the symmetric sparse matrix-vector multiplication in multicore. In 27th IEEE International Parallel & Distributed Processing Symposium (IPDPS'13), Boston, MA, USA, 2013. IEEE. [pdf], [pub]
  • V. Karakasis, G. Goumas, K. Nikas, N. Koziris, J. Ruokolainen, and P. Råback. Using State-of-the-Art Sparse Matrix Optimizations for Accelerating the Performance of Multiphysics Simulations. In PARA 2012: Workshop on State-of-the-Art in Scientific and Parallel Computing, Helsinki, Finland, 2012. Springer. [pdf], [pub]
  • V. Karakasis, G. Goumas, and N. Koziris. Exploring the performance-energy tradeoffs in sparse matrix-vector multiplication. In Workshop on Emerging Supercomputing Technologies (WEST), ICS'11, Tucson, AZ, USA, 2011. [pdf]
  • K. Kourtis, V. Karakasis, G. Goumas, and N. Koziris. CSX: An extended compression format for SpMV on shared memory systems. In 16th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming (PPoPP'11), San Antonio, TX, USA, 2011. ACM. [pdf], [pub]
  • V. Karakasis, G. Goumas, and N. Koziris. Performance models for blocked sparse-matrix-vector multiplication kernels. In 38th International Conference on Parallel Processing (ICPP'09), Vienna, Austria, 2009. IEEE Computer Society. [pdf], [pub]
  • V. Karakasis, G. Goumas, and N. Koziris. A comparative study of blocking storage methods for sparse matrices on multicore architectures. In 12th IEEE International Conference on Computational Science and Engineerging (CSE-09), Vancouver, Canada, 2009. IEEE Computer Society. [pdf], [pub]
  • V. Karakasis, G. Goumas, and N. Koziris. Exploring the effect of block shapes on the performance of sparse kernels. In 10th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-09), IPDPS'09, Rome, Italy. IEEE Computer Society. [pdf], [pub]
  • G. Goumas, K. Kourtis, N. Anastopoulos, V. Karakasis, and N. Koziris. Performance evaluation of the sparse matrix-vector multiplication on modern architectures. The Journal of Supercomputing, 50(1):36–77, 2009. Springer. [pdf], [pub]
  • G. Goumas, K. Kourtis, N. Anastopoulos, V. Karakasis, and N. Koziris. Understanding the performance of sparse matrix-vector multiplication. In Proceedings of the 16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP'08), Toulouse, France. IEEE Computer Society. [pdf], [pub]
  • G. Goumas, N. Drosinos, V. Karakasis and N. Koziris. Coarse-grain parallel execution for 2-dimensional PDE problems. In 8th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-07), IPDPS'07, Long Beach, CA, USA, 2007 IEEE Computer Society. [pdf], [pub]

publications.isys.list

  • V. Karakasis and A. Stafylopatis. Efficient evolution of accurate classification rules using a combination of gene expression programming and clonal selection. IEEE Transactions on Evolutionary Computation, 12(6):662–678, 2008. Springer. [pdf], [pub]
  • A. Lanaridis, V. Karakasis, and A. Stafylopatis Clonal selection-based neural classifier. In Eighth International Conference on Hybrid Intelligent Systems (HIS'08), 2008, Barcelona, Spain. IEEE. [pdf], [pub]
  • V. Karakasis and A. Stafylopatis Data mining based on gene expression programming and clonal selection. In IEEE Congress on Evolutionary Computation (CEC 2006), Vancouver, Canada, 2006. IEEE. [pdf], [pub]