Computational engineering and science have mainly addressed computer modeling and simulation to create surrogates of real entities. Accurate representations usually require considerable amounts of computer power and data access. Software applications developed in the past can no longer adequately support the emerging highly parallel computing devices, complemented by different accelerator devices with higher levels of parallelism, such as the CUDA/GPU or the many-core Xeon Phi.

Current computer clusters are highly heterogeneous: inter-connected nodes with distributed memories, where each node supports several computing units that share part of the memory hierarchy and is coupled with distributed memory accelerators.

How can we clearly identify and overcome these challenges to build performance portable applications?

What has been our experience dealing with the development/deployment of scientific applications that face these challenges?

Topics of interest include but are not limited to:



The proceedings of the HPCES 2014 Session will appear in the Springer-Verlag Lecture Notes in Computer Science (LNCS) series.


The paper must be camera-ready, between 10 and 16 pages long and formatted according to the LNCS rules (

To submit a paper, please connect to the submission page ( from the link available at the ICCSA 2014 web site. Only papers submitted through the electronic system and strictly adhering to the relevant format will be considered for reviewing and publication.

The paper must deal with original and unpublished work, not submitted for publication elsewhere. All submissions will be reviewed by at least three experts in the relevant field.

Please pay attention, when submitting your contribution to select the right entry “High Performance Computing in Engineering and Science - HPCES 2014” in the listbox shown in the submission form.



Alberto Proenca (aproenca<at>
  Dep. Informática, EEUM
  Universidade do Minho
   Pedro Alberto (pedro.alberto<at>
  Dep. Física, FCTUC
  Universidade de Coimbra