AMRMMHD: Adaptive multi-resolution massively-multicore hybrid dynamics
New algorithms and architectures are needed to exploit the next generation of heterogeneous, massively parallel processors (such as those found in widely available graphics processors:GPUs) to deliver orders-of-magnitude performance increases for conformational sampling in molecular simulations.
The AMRMMHD project is taking the existing Sire molecular modelling code and adding a new adaptive, hybrid multiresolution dynamics program as well as a workpacket scheduling system. This allows it to target new architectures to explore rational drug discovery by modelling protein conformational change and protein flexibility.
Sire is a multiscale molecular simulation framework, written to allow computational modelers to quickly prototype and develop new algorithms for molecular simulation and molecular design. Sire is written as a collection of libraries, each of which contains self-contained and robust C++/Python building blocks. These building blocks are thread-aware and can be streamed (saved/loaded) to and from a version-controlled and tagged binary format, thereby allowing these building blocks to be combined together easily to build custom multi-processor molecular simulation applications.
Sire is developed by the Centre for Computational Chemistry at the University of Bristol, who are working with the Software Sustainability Institute to scale the software development process from a single developer to a small development team spread across multiple sites.
For further information visit the Sire website.