Slicast
Saturday, October 31, 2026
DarkSubscribe
Slicast
AI Infrastructure · News & Analysis
Home › Chips & Hardware › Report
Chips & Hardware · Report

A feature argues that switches are not scaling fast enough for AI data centers, a bottleneck that could hurt data center operators.

If switch bandwidth and port scaling lag GPU cluster growth, operators face stranded compute and slower cluster builds, making networking a gating constraint alongside power and chips.
Trade pressSlicast · October 30, 2026 at 15:11 UTC · Global · Source: Data Center Dynamics
importance 50
Read the original
Related reports
Chips & HardwareBlocks & Files says Samsung's memory business is virtually printing money, with the source giving no figures.2026-10-31Chips & HardwarePsiQuantum has turned its photonics know-how into an eight-port optical switch for AI data center GPU networks, typically losing 1.2 dB of light, versus the up to 3 dB loss that data centers allow.2026-10-31Chips & HardwareIntel shares fell roughly 25–29% in a single session, on course for their worst day on record, after a weak outlook and a PC market slump overshadowed an earnings beat. The sell-off spread into a global chip rout, with Dell, AMD and Nvidia also declining. Reports differ on the exact size of the drop.2026-10-31Chips & HardwareOpenAI CEO Sam Altman said on a podcast that it is 'quite terrible' for people to keep RAM in their own PCs when that memory could go into AI data centers, which he argued offer better utilization.2026-10-31Chips & HardwareAmazon's custom silicon, including Trainium and Graviton, reportedly posted a 20x revenue surge in the third quarter, according to The Tech Buzz.2026-10-31
A feature argues that switches are not scaling… · Slicast