Developer Reference

Contents

Graph Routines

Intel® oneAPI Math Kernel Library (oneMKL)
provides a set of routines to perform various operations on graphs as represented in the language of linear algebra. The functionality is currently added in a preview mode. In this preview mode, the library provides only partial support for some of the documented functionality. For instance,
mkl_graph_mxm
and
mkl_graph_mxv
support different subsets of the full list of semirings. Supported configurations are documented for each routine, and routines will return
MKL_GRAPH_STATUS_NOT_SUPPORTED
when the input configuration is not supported. As a best practice, always check for successful return status (
MKL_GRAPH_STATUS_SUCCESS
) from all calls to the MKL graph routines. More functionality will be added in future releases. The API is based on the GraphBLAS C API specification, but is less general and has a more limited scope. Comments, questions and suggestions are encouraged and should be submitted to the
Intel® oneAPI Math Kernel Library (oneMKL)
.
The terms and concepts required to understand the use of the
Intel® oneAPI Math Kernel Library
graph routines are discussed in Appendix F: Graph Basics. If you are familiar with GraphBLAS terminology and sparse matrix storage schemes, you can skip these sections and go directly to the interface descriptions.
Refer to the description of
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.
Notice revision #20201201

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.