Accelerating Scientific Computations with Mixed Precision Algorithms

Abstract

On modern architectures, the performance of 32-bit operations is often at least twice as fast as the performance of 64-bit operations. By using a combination of 32-bit and 64-bit floating point arithmetic, the performance of many dense and sparse linear algebra algorithms can be significantly enhanced while maintaining the 64-bit accuracy of the resulting solution. The approach presented here can apply not only to conventional processors but also to other technologies such as Field Programmable Gate Arrays (FPGA), Graphical Processing Units (GPU), and the STI Cell BE processor. Results on modern processor architectures and the STI Cell BE are presented.

Cited by in corpus (17)

• Majorana Braiding Dynamics on Nanowires
• Reproducibility, accuracy and performance of the Feltor code and library on parallel computer architectures
• Numerical integration on GPUs for higher order finite elements
• Enabling scalable stochastic gradient-based inference for Gaussian processes by employing the Unbiased LInear System SolvEr (ULISSE)
• A directive based hybrid Met Office NERC Cloud model
• Mixed-precision for Linear Solvers in Global Geophysical Flows
• Compression challenges in large scale PDE solvers
• Profile-Driven Automated Mixed Precision
• A mixed precision semi-Lagrangian algorithm and its performance on accelerators
• Leveraging Mixed Precision in Exponential Time Integration Methods
• A note on solving nonlinear optimization problems in variable precision
• Real-time adaptive sensing of nuclear spins by a single-spin quantum sensor
• An implicit symplectic solver for high-precision long term integrations of the Solar System
• GPU Acceleration of Real-Time Control Loops
• Iterative diagonalization of symmetric matrices in mixed precision
• A multigrid reduction framework for domains with symmetries
• An adaptive regularization algorithm for unconstrained optimization with inexact function and derivatives values