A load-balancing workload distribution scheme for three-body interaction computation on Graphics Processing Units (GPU)

• Load-balancing scheme for calculating three-body interactions on GPU.
• Perfect load-balancing is achieved if NN is not divisible by 3.
• Nearly perfect load-balancing is obtained if NN is divisible by 3.
• Parallel efficiency demonstrated in three-body potentials.

Three-body effects play an important role for obtaining quantitatively high accuracy in a variety of molecular simulation applications. However, evaluation of three-body potentials is computationally costly, generally of O(N3N3) where NN is the number of particles in a system. In this paper, we present a load-balancing workload distribution scheme for calculating three-body interactions by taking advantage of the Graphics Processing Units (GPU) architectures. Perfect load-balancing is achieved if NN is not divisible by 3 and nearly perfect load-balancing is obtained if NN is divisible by 3. The workload distribution scheme is particularly suitable for the GPU’s Single Instruction Multiple Threads (SIMT) architecture, where particle’s data accessed by threads can be coalesced into efficient memory transactions. We use two potential energy functions with three-body terms, the Axilrod–Teller potential and the Context-based Secondary Structure Potential, as examples to demonstrate the effectiveness of our workload distribution scheme.