Abstract
In this paper, we present a new parallel computation model for realistic rendering and 3D simulation applications. Our major goal is to minimize the programmer's work in order to develop, to obtain and to maintain maximal efficiency of this parallel application. To present the framework of our work, we describe the applications we aim at and we explain why object-oriented programming and SPMD (Simple Program Multiple Data i.e. all the processors run the same code in an asynchronous way on different data.) seem to be well suited to the hardware target platform. Then, after a short survey of some existent approaches, we detail our proposition, based on a data distribution facility set offered through a general SPMD programming structure. First, we show how a classical object-oriented language improved by some special concurrent aggregate classes can be sufficient. Then we study how some techniques of virtual communication can efficiently les us manage these features. We end with a discussion as regards implementation problems on our present simulator running on a T800 Transputer network and for the design on a future T9000 Transputer machine.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
