· ai-devtools

AMD: Where we're going we don't need (as much) DRAM

AMD's latest sales pitch: Up to 99 percent of the performance with half the DRAM, provided you're willing to buy a Mext license alongside your shiny new Epyc server.


This is made possible by using AI models to load memory pages in and out of cheaper NVMe storage before your operating system or apps ever know they need it — kind of like SWAP or a Windows page file, but smarter. AMD acquired the tech from Mext back in June, and according to former CEO Gary Smerdon — now CVP of business development at AMD — the tech could cut enterprise server costs by as much as 83 percent by replacing costly RDIMMs with flash storage. Of the initial 145 systems the software has been deployed on so far, he estimates customers have already saved $5 million in avoided DRAM costs.

Earlier coverage

Related coverage

Also worth reading

  1. angular 22.3.0-rc.0 released
  2. node 26.13.1 released
  3. The Rise of AI Site Reliability Engineers – Can Autonomous Agents Handle Production Incidents
  4. Red Hat OpenShift 5 brings autonomous troubleshooting to hybrid clouds
  5. Druva Agent Recovery restores AI agents, supports Copilot Studio, Power Apps, Power Automate