oneMath

The oneAPI Math Library (oneMath) defines a set of fundamental mathematical routines for use in high-performance computing and other applications. As part of oneAPI, oneMath is designed to allow execution on a wide variety of computational devices: CPUs, GPUs, FPGAs, and other accelerators. The functionality is subdivided into several domains: dense linear algebra, sparse linear algebra, discrete Fourier transforms, random number generators and vector math.

The general assumptions, design features and requirements for the oneMath library and host-to-device computational routines will be described in oneMath Architecture. The individual domains and their APIs are described in oneMath Domains. Other design considerations that are not necessarily part of the oneMath specification but that are worth mentioning will be discussed in oneMath Appendix.

• oneMath Architecture
  • Execution Model
  • Memory Model
  • API Design
  • Exceptions and Error Handling
  • Other Features
• oneMath Domains
  • Dense Linear Algebra
  • Sparse Linear Algebra
  • Discrete Fourier Transforms
  • Random Number Generators
  • Summary Statistics
  • Vector Math
• oneMath Appendix
  • Future considerations
  • Acknowledgment