← All stories

Rivento topic desk

Interface Silicon

Focused coverage of the tools, architectures, and manufacturing decisions moving this part of the AI silicon stack forward.

Marvell
IEEE Spectrum · 2026-09-02

CXL 3.1 Memory Pooling Controllers Unlock Low-Latency Fabric Expansion for AI Data Centers

Silicon developers have taped out advanced Compute Express Link controllers that enable pooled server memory sharing across rack-scale fabrics with near-native DRAM latency.

Read briefing