KTH Royal Institute of Technology


KTH Royal Institute of Technology
Address:  KTH, SE-10044, Stockholm, Sweden
Country: Sweden
kth
PDC Center for High Performance Computing
Director: Prof. Erwin Laure
Address: PDC, KTH, SE-10044, Stockholm,Sweden
 kth2

Institution Background

KTH Royal Institute of Technology in Stockholm is the largest and oldest technical university in Sweden. No less than one-third of Sweden’s technical research and engineering education capacity at university level is provided by KTH. Education and research spans from natural sciences to all branches of engineering and includes Architecture, Industrial Management and Urban Planning. There are a total of 13,400 first and second level students and almost 1,900 doctoral students. KTH has 4,900 employees.

 

Research Group Background

The PDC Center for High-Performance Computing at the KTH Royal Institute of Technology is the leading provider of HPC services for academic research in Sweden. PDC is located centrally in Stockholm under the auspices of the KTH CSC School of Computer Science and Communication. PDC’s services are made available to Swedish and European researchers, respectively via the Swedish National Infrastructure for Computing (SNIC) and PRACE. PDC also provides its services to the commercial sector via collaborative projects. The main HPC system currently running at PDC is Lindgren, a 305 teraflop Cray XE6, however planning is underway to replace this with a new system with a capacity in the petaflops range within the next year.

PDC is an active partner in many national and international research projects. The systems at PDC are integrated with pan-European collaborative e-infrastructures such as PRACE and EGI, as well as with local and international cloud facilities. Through these collaborative projects, PDC (together with other European supercomputing centres) is active in world-leading research in high-performance computing.

 

Major Interest in Action Scientific Topics

As we look forward to the imminent arrival of exascale supercomputers on the research scene, new programming languages and parallelization techniques will be needed to help e-Science applications run efficiently on systems containing billions of computing units. PDC, as a member of the exascale project CRESTA, is leading efforts to investigate new programming models and parallelization techniques for exascale computing in e-Science application areas such as computational fluid dynamics simulations of highly complex flows (like air flowing around airplane wings and fluid flows inside pipes).

PDC is actively investigating new programming languages and parallelization strategies to fully exploit future exascale computing platforms. Exascale supercomputers may have heterogeneous architectures including CPU, GPU and cell accelerators.  The new programming languages/libraries under investigation are installed on PDC supercomputers, and are available for testing purposes and also for developing new computer codes.

PDC’s research into new programming models for high performance computing includes the study of mixed GPU/CPU programming with compiler directives, GPU programming with CUDA and Partitioned Global Adress Space (PGAS) languages.