Porting Math Libraries Across Heterogeneous Architectures
Existing math kernel libraries have lacked portability across heterogeneous platforms—until now. A unifying programming model—and availability of standard library interfaces—enables...
Audio is streamed directly from the publisher (media20.connectedsocialmedia.com) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
Existing math kernel libraries have lacked portability across heterogeneous platforms—until now. A unifying programming model—and availability of standard library interfaces—enables development of performance-portable libraries among diverse hardware architectures. In this episode, Julia Sukharina, Senior Engineering Manager at Intel, and Mehdi Goli, Principal Software Engineer at Codeplay Software, talk about a collaborative project to make this portability possible, and Mehdi dives into his work to enable the first math library implementation for #oneAPI on Nvidia GPUs. Have a favorite library you want to enable, or want to contribute a proposal for a new math interface to drive standardization across multiple hardware?
To get started, visit:
github.com/oneAPI-SRC/oneMKL
Read/Download the transcript