BALANCING METHOD FOR PARTITIONED SOLVERS

Image

It is my distinct honour and privilege to welcome you to the Journal of Theoretical and Computational Science.

The Journal of Theoretical and Computational Science aims to spread knowledge and promote discussion through the publication of peer-reviewed, high quality research papers on all topics related to Modern Scientific Techniques. The open access journal is published by Longdom Publishing who hosts open access peer-reviewed journals as well as organizes conferences that hosts the work of researchers in a manner that exemplifies the highest standards in research integrity.

 Performance Model Normal Form (PMNF) regression is considered to find an empirical performance model for each solver. Then, an appropriate optimization problem is derived and solved to find the optimal core distribution between solvers. The optimization problem directly depends on the equation coupling type (serial or parallel).

In addition, due to the optimal usage of computation capacity, the new method considerably improves the scalability. We also compare the load balancing results with a solver-specific scheme and show that, even though the data-based method does not explicitly use information about mesh size, discretization type and numerical methods used by solvers, it can achieve comparable results.

We always encourage your research works under the scope of our Journal of Theoretical and Computational Science. (Tap on the link to submit your research work)

Warm Regards

Angelina

Managing Editor

Journal of Theoretical and Computational Science

WhatsApp: +3225889658