Device Memory TCP Nears The Finish Line For More Efficient Networked Accelerators

Posted by Scott_Ruecker on Jul 9, 2024 11:35 AM EDT
Phoronix; By Michael Larabel 		Mail this story
Print this story

A year ago Google engineers posted experimental Linux code for Device Memory TCP for more efficient transferring of data from GPUs/accelerators to network devices without having to go through a host CPU memory buffer. After going through many rounds of review, Device Memory TCP appears to be nearing the finish line.

Full Story

  Nav
» Read more about: Story Type: News Story; Groups: Linux

« Return to the newswire homepage

This topic does not have any threads posted yet!

You cannot post until you login.