Parallel Processing Laboratory


Parallel computing is the simultaneous execution of an application that is split up into many smaller, coordinated tasks on multiple processors in order to obtain results faster. In general parallel computing involves algorithms, programming models, hardware, and applications.

A parallel computing system is either a set of computers with a high-speed network connection (cluster) or a computer with more than one processor (parallel computer). Multicore, manycore, GPGPU, FPGA, grid and cloud computing systems are also suitable platforms for parallel computing.

Current research interests of the lab members are as follows:

  • High performance computing
  • Development of parallel algorithms and their applications in science and engineering
  • Parallel sparse matrix algorithms
  • Parallel application development and run-time environments

Related Computing Resources


  • Ercan Selçuk Bölükbaşı
  • Erdem Sarıgil
  • Edona Fasllija
  • Hamide Hande Keskiner
  • Hilal Kılıç
  • Murat Manguoğlu, Assoc. Prof.
  • Süleyman Salın
  • Cevat Şener, Dr.
  • Orkun Tanık


  • “Parallel Solution of Sparse Linear Systems”, TUBITAK Career Award (EEAG111E238), (2012-2014)
  • “Solution of Linear Systems on Parallel Computers”, BAP-08-11-2011-128, (2011-2013).
  • “Academic Link Development between Queen Mary and Westfield College of London University and METU for Joint Research on Parallel Computing” under the program “Academic Link Scheme for Turkey”, supported by British Council (1993-1997).
  • TÜBİTAK Research Project, “Development of Parallel Computing Library and Their Benchmarks on a Cluster”, EEEAG102E010, (2002-2004).
  • “Parallel Computing Laboratory based on a Transputer System”, BAP, (1989).
  • “Construction of an Algorithm Library by Developing Parallel Cluster Management and Monitoring Tool”, BAP-2002.03.12.03, (2002).

Recent Publications

  • Uzunca M, Karasozen B, Manguoglu M,Adaptive discontinuous Galerkin methods for nonlinear diffusion-convection-reaction equations, Computers and Chemical Engineering, 63, pp.24-37, 2014
  • Manguoglu M, Parallel solution of sparse linear systems,High Performance Scientific Computing, Springer (book chapter, editors:M. Berry, K. Gallivan, E. Gallopoulos, A. Grama, B. Philippe, Y. Saad and F. Saied), 2012
  • Manguoglu M, A domain-decomposing parallel sparse linear system solver, Journal of Computational and Applied Mathematics, 236(3), 319-325, 2011
  • Manguoglu M., Takizawa K., Sameh A., Teduyar T., A parallel sparse algorithm targeting arterial fluid mechanics computations, Computational Mechanics, 48(3), 377-384, 2011
  • Dag H., Yetkin F., Manguoglu M., A New Preconditioner Design Based on Spectral Division for Power Flow Analysis, International Review of Electrical Engineering, 6(3), pp.1339-1348, 2011
  • Manguoglu M., Saied F., Sameh A., Grama A., Performance Models for the Spike Banded Linear System Solver, Scientific Programming, 19(1),pp.13-25, 2011
  • Brodman J.C., Evans G.C., Garzaran M.J., Manguoglu M., Sameh A., Padua D., A Parallel Numerical Solver Using Hierarchically Tiled Arrays, Lecture Notes in Computer Science(proceedings of LCPC10), Volume 6548, pp.46-61, 2011
  • Manguoglu M., Cox E., Saied F., Sameh A., TRACEMIN-Fiedler: A Parallel Algorithm for Computing the Fiedler Vector, Lecture Notes in Computer Science(proceedings of VECPAR10), Volume 6449, pp.449-455, 2011
  • Şener, C., Paker, Y. and Kiper, A., Developing a data-parallel aplication with DaParT, Proceedings of PPAM'2001: Fourth International Conference on Parallel Processing and Applied Mathematics, Naleczow, Poland, September 2001.
  • Şener, C. DaParT: A Data-Parallel Programming Tool, Ph.D. Thesis, Department of Computer Engineering, METU, January 2000.
  • Kiper, A,. A Parallel Approach for Quadratic Eigenproblems, Proceedings of the 16th IMACS World Congress on Scientific Computation, Applied Mathematics and Simulation (Editors: Michel Deville and Robert Owens), Switzerland, August 21-25, 2000.
  • Kiper, A., A Parallel Triangular Inversion Using Elimination, Abstracts of the International Workshop on Parallel Matrix Algorithms and Applications, Switzerland, August 17-20, 2000.


If you are interested in doing research in the area of parallel and/or scientific computing the following is a list of recommended courses offered in our department.

  • CENG371 - Scientific Computing (*)
  • CENG478 - Introduction to Parallel Computing (*)
  • CENG493 - Special Topics in Comp. Eng: Cluster Computing (*)
  • CENG571 - Numerical Analysis I
  • CENG572 - Numerical Analysis II
  • CENG576 - Numerical Methods in Optimization
  • CENG577 - Parallel Computing
  • CENG780 - Sparse Matrix Computations (NEW)

(*) Master's students can also register to at most two undergraduate level courses as technical electives provided they have not taken those courses during their undergraduate studies.

Former Members

  • Prof. Dr. Ayşe Kiper
  • Mehmet Gülek
  • Cüneyt Mertayak
  • Murat Gençtav
  • Meftun Cincioğlu
  • Ömer Tarı
research/parallel/index.txt · Last modified: 2014/10/14 09:00 by Murat Manguoglu