As a technology with the potential to influence computing and society as a whole, Parallware is novel LLVM-Based Software Technology for Classification of Scientific Codes to Assist in Parallelization with OpenMP and OpenACC.
Parallware is a new technology for static analysis of programs based on the production-grade LLVM compiler infrastructure. Using a fast, extensible hierarchical classification scheme to address dependence analysis, it discovers parallelism and annotates the source code with the most appropriate OpenMP & OpenACC directives. Published success stories have already shown the potential of the new technology with microbenchs and with the NAS Parallel Benchmark EP, covering fields such as finite elements, computational electromagnetics and sparse codes. Thanks to this, we will receive an exhibit floor space in a high visibility location; this location will allow attendees the opportunity to witness our technology demonstrations, see presentations, and hold in-depth technical discussions. We will meet you there.
Parallware tools offer a new parallel programming environment that helps to manage the complexity of developing parallel programs for large HPC facilities. Currently, the portfolio of Parallware tools under construction is: Parallware Trainer, a desktop tool for effective HPC training; and Parallware Assistant, desktop tool for high productivity guided parallelization to simplify the effort of HPC developers.
“Our goal is to first present the new tool Parallware Trainer at SC16. Later on, we will develop additional modules for the Parallware Assistant during 2017 so that HPC experts have access to detailed information about their scientific programs, as well as control on the paradigm used to generate parallel code (see to the right “threading”, “offloading”, “SIMD”).”
In this video, Appentra CEO Manuel Arenaz describes Parallware.
In related news, registration is now open for the third annual StartupHPC Confererence, which will kick off on Monday, Nov. 14 in Salt Lake City.