Welcome! I am currently working as a PostDoc researcher at KAUST. My main research area is parallel scientific computing. I have expertise on:
- data partitioning and task mapping/scheduling for exascale applications
- locality aware partitioning and scheduling of irregular applications
- parallel programming (on IBM BlueGene/P,Q and other parallel systems)
- cache-aware methods and programming for scientific computational kernels
Publications:
- Hypergraph Partitioning Based Models and Methods for Exploiting Cache Locality in Sparse Matrix-Vector Multiplication Akbudak K., Kayaaslan E., and Aykanat C., SIAM Journal on Scientific Computing 2013 35:3, C237-C262
- Simultaneous Input and Output Matrix Partitioning for Outer-Product-Parallel Sparse Matrix-Matrix Multiplication, Akbudak K. and Aykanat C., SIAM Journal on Scientific Computing 2014 36(5), C568–C590
- Locality-Aware Parallel Sparse Matrix-Vector and Matrix-Transpose-Vector Multiplication on Many-Core Processors, Karsavuran O., Akbudak K, and Aykanat C., IEEE Transactions on Parallel and Distributed Systems ( Volume: 27, Issue: 6, June 1 2016 )
- Exploiting Locality in Sparse Matrix-Matrix Multiplication on Many-Core Architectures, Akbudak K, and Aykanat C., IEEE Transactions on Parallel and Distributed Systems (accepted)
Invited Talk:
- K. Akbudak, R. O. Selvitopi, and C. Aykanat, Exploiting Sparsity in Parallel Sparse Matrix-Matrix Multiplication, SIAM Conference on Applied Linear Algebra (LA15), October 26-30, 2015, Atlanta
Workshops:
- O. Karsavuran, K. Akbudak, and C. Aykanat, Exploiting Matrix Reuse and Data Locality in Sparse Matrix-Vector and Matrix-Transpose-Vector Multiplication on Many-Core Architectures, SIAM Workshop on Combinatorial Scientific Computing (CSC), Albuquerque, NM, USA, October 10-12, 2016
Projects:
- PRACE-1IP: PRACE (Partnership for Advanced Computing in Europe) First Implementation Phase Project (2010 - 2012), European Union - FP7, Grant Agreement No: RI-261557, Bilkent Budget: 205,000 Euro Task 7.5 - Programming Techniques for High Performance Applications Subtask 7.5A&B - Analysis of Scalable Algorithms Task 7.1 - Internal Call Improving sparse matrix multiplication operations in CP2K
- PRACE-1IP-EXT 250K core-hour granted on JUQUEEN as computing resource award by PRACE (Preparatory Access project 2010PA0930 awarded at 9th cut-off) 250K core-hour granted on JUQUEEN as computing resource award by PRACE (Preparatory Access Type B project 2010PA2155 awarded at 15th cut-off, 02/12/2013)
- PRACE-4IP Task 4.3: On-line Training Subtask: Evaluation of platforms for the CodeVault Task 7.2: Preparing for Future PRACE Exascale Systems
Journal/Conference Referee Services:
- Parallel Computing (ParCo)
- Journal on Parallel and Distributed Computing (JPDC): 2012
- IPDPS 2013
- IPDPS 2014
- EuroPar 2014
- IPDPS 2015
- IPDPS 2016
- SOFSEM-FOCS2017
My Web Sites: