Beta
Subscribe

Memory Architecture

Articles (3)

A triangular silicon wafer chip lifted from a stone bowl of red salsa, molten sauce dripping from its edge.
AInews

At Hot Chips 2026, the Peak-FLOPS Race Was Really a Contest to Keep Compute Fed

Compute scales about 3x every two years; memory bandwidth scales under 2x. NVIDIA, AMD, Google, Intel, and OpenAI each answered that gap differently.

Advanced PackagingSemiconductor Manufacturing
SCN Staff·
Glowing coil of optical fiber encircling a row of dark data-center server racks, green pulses of light circulating through the glass.
Emergingnews

Delay-Line Memory Returns: A Swedish Team Proposes Trading HBM for 14,000 Kilometers of Glass

Uppsala researchers propose streaming LLM weights through spooled fiber instead of HBM. The design is a sketch; the DRAM market that provoked it is very real.

AI InfrastructureMemory Architecture
SCN Staff·
A SOCAMM2 LPDDR5X server memory module showing dense chip array architecture, illuminated against a black background, designed for next-generation AI server platforms.
AIanalysis

Vera Rubin's Memory Stack Is Korean. How Three Vendors Got There Tells You Why It Will Stay That Way.

Samsung, SK hynix, and Micron converged on SOCAMM2 mass production within six weeks for NVIDIA's Vera Rubin. Korean suppliers now control both memory tiers.

NVIDIASemiconductor Manufacturing
SCN Staff·