As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Object-oriented (OO) programming conquered mainstream in the 1990s but the procedural programming languages Fortran and C are still dominating in the High-Performance-Computing (HPC) community. However in recent years, compute kernels or even whole applications have been migrated from C to object-oriented C++ and Fortran has been retrofitted with OO capabilities as well. In this work we illustrate how OO abstractions can be used to introduce Shared-Memory parallelization into existing applications by using the concepts of encapsulation and modularity. For the important domain of sparse linear algebra routines, we discuss several parallel implementation strategies and data type design considerations. We aim to maximize performance and scalability while hiding the implementation details as much as possible.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.