Multi-core Optimization Program

Multi-core technology is a reality of today. The era of the single processor system has passed; multi-core is real as applications can no longer count on increased processor clock speeds to improve performance. Historically, the majority of applications are based on a single-threaded architecture, and dependent on clock speeds to increase performance. They gain no linear performance improvements when run on multi-core systems if they have not been designed to take advantage of the increased multiple compute engines available on the chips. If the multi-core aspect of systems is not taken into consideration, today’s applications may run slower than even with more compute power available in the system.

With the focus on hardware, and increasing processor counts, there is increasing need to understand the new complexities of application design, debug, and optimization in multi-core systems. In order to take advantage of the additional processing power that multi-core systems offer, new development tools are needed that allow the applications to change as well. Exploiting the power of multi-core processors will be critical for customers to improve their business success.

The goal of the Multi-core Program is to optimize HPC applications on multi-core systems so that customers can realize the full benefits of the technology faster. The program brings together a robust portfolio of products and technologies from HP and our development partners. These technologies include in-house products, software tools from our partners and those developed in collaboration with our customers.

Our extensive development tools portfolio, including compilers, debuggers, math libraries, performance tuning and other development tools, are now available for a variety of processor and system architectures, operating systems and interconnects. Our expanded portfolio, packaged as a multi-core application development suite, is available now.