50%: AMD's new software lets Epyc servers use half the memory
Mext moves rarely used data to flash and predicts what will be needed to bring it back, with a small performance loss.
Server memory is expensive. New AMD software moves data that is rarely used to a cheaper place and brings it back when needed. Companies then need about half the memory, without a big slowdown.
Hype check
The noise matches the real change.
Our editorial opinion, based on the sources below. How we score
Your call: hype or real?
Real = it will matter in daily use. Hype = louder than it is. One vote, and it can’t be changed.
Our view and readers’ views are opinions, not statements of fact about a product.
Why it matters
Server memory is very costly, and using less may lower cloud costs over time.
Skip: news for data centers, not your device.
The details
- AMD acquired Mext in June; its software runs on Epyc servers.
- It moves rarely used memory pages from DRAM to cheaper flash and uses machine-learning models to bring them back before they are needed.
- In Synopsys PrimeTime, 95% of performance remained after replacing half the DRAM.
- In Apache Spark, 99% of performance remained with half the DRAM.
- It already runs on 145 systems, and AMD is preparing an enterprise release with partners such as HPE.
- Pricing and a wider launch date are not announced, and it works only with AMD hardware.
Receipts
One trusted outlet reports it. Not confirmed yet. Open the sources and check us.
Written with the help of AI tools and checked against the sources above. How we work
See a mistake? Tell us and we’ll fix it.
