The EU unveiled a tech sovereignty package to cut reliance on foreign cloud, semiconductor, and software providers via new rules, procurement shifts, and chip production powers. 🇪🇺
It adds cloud sovereignty tiers and favors EU-aligned providers for sensitive public services while keeping most markets open. ⚙️

🔗 https://www.techpolicy.press/eu-unveils-sweeping-tech-sovereignty-push-balancing-autonomy-with-openness/

#TechNews #EU #TechSovereignty #CloudComputing #Semiconductors #AI #OpenSource #Infrastructure #Privacy #Security #Technology #Policy #Europe #DataSovereignty

Intel's fab roadmap examined — Arizona, Ohio, Ireland, and the two deadlines deciding 14A process node

https://fed.brid.gy/r/https://www.tomshardware.com/tech-industry/semiconductors/intels-fab-roadmap-examined

Fed held at 3.50-3.75% today, unanimous. Warsh's first FOMC as chair, no cut signal.

Yet semis are outperforming: $AMAT and $LRCX green while the S&P and Nasdaq slip. AI capex over the Fed, so far.

Rotation for real, or a one-day fade? What's your read?

#FOMC #Fed #StockMarket #Semiconductors

AMD bought MEXT, a startup whose software makes cheap NAND flash act like DRAM, so a server runs on far less of the expensive stuff. With DRAM contract prices up ~90%+ this year, the logic isn't subtle. How predictive memory tiering actually works:
https://www.buysellram.com/blog/amd-mext-flash-as-dram/

#AMD #MEXT #DRAM #NANDFlash #MemoryTiering #AIInfrastructure #DataCenter #ServerHardware #SSD #ITAD #Semiconductors #HBM

AMD just spent money to make memory cheaper, not faster. Its acquisition of MEXT, a small software company, is a bet on a simple idea: most of the DRAM in a server sits idle, and cheap NAND flash can stand in for it if the software is smart enough.

MEXT's Predictive Memory Engine watches how an application uses memory, predicts which pages are about to be needed, and moves them from flash up to DRAM before the program asks. Hot data stays fast; cold data lives on flash that costs a fraction as much. The application never notices.

Conventional DRAM contract prices jumped roughly 93-98% in a single quarter as makers shifted capacity to HBM for AI accelerators. When DRAM is this scarce, stretching it with flash stops being a niche trick.

One caveat: flash being ~50x cheaper per bit doesn't mean a 50x cut to a memory bill. The realistic payoff is closer to half the DRAM cost. Still, the direction is clear, and it changes what aging server memory and SSDs are worth on the way out.

https://www.buysellram.com/blog/amd-mext-flash-as-dram/
#AMD #MEXT #DRAM #NANDFlash #MemoryTiering #AIInfrastructure #DataCenter #ServerHardware #SSD #ITAD #Semiconductors #HBM #tech