Microsoft reveals Maia 200 & network architecture pushing scale-up inference efficiency
Hyperscaler leverages a two-tier Ethernet-based topology, custom AI Transport Layer & software tools to deliver a tightly integrated, low-latency platform
Hyperscaler leverages a two-tier Ethernet-based topology, custom AI Transport Layer & software tools to deliver a tightly integrated, low-latency platform
HBM4 win for Korean giant as it finds parity with SK Hynix
First RISC-V company to partner with Nvidia on the interconnect technology
2026 projected to be an SSD dry year, especially with the boon of key value caching
First memory chips. Then lasers. Now, copper looks to be the next commodity to pose ‘systemic risk’ to the digital infrastructure space
CES 2026 showcase sees vendor join the growing rush to market with next-generation network processors years ahead of the standard's finalization
Second acquisition in a month would see PCIe and CXL switch expert follow Celestial AI
AWS doubles down on custom silicon and the looming memory crisis to come
Korean giant also working with SanDisk
Dell dominates both markets, but only one side is throwing a party
Samsung meanwhile looks to be on Nvidia's good list for HBM4
Optical module makers could face component shortage as Nvidia reportedly hoards EML supply
UChicago researchers’ novel nanofabrication method supercharges quantum coherence
HBM4 memory reportedly pushed back into latter half of 2026
Update to openSUSE brings artificial intelligence closer to home
Not all growth would be equal, however, with Europe significantly lagging behind Asia and the U.S.
Non-exclusive deal aims to accelerate chip-design workflows
Despite hobbled hardware, DeepSeek bypasses bandwidth constraints with 'sparse attention' design
Deal would give Marvell access to Celestial’s pipeline of photonic interconnect technology
AI driving firm's latest company restructure plan
Custom hardware delivers 73% more bandwidth while cutting power consumption by 35%
Gen 8 Fiber Channel switches use updated crypto algorithms, not quantum mechanics to ward off future threats
Neoverse integration means custom hardware from AWS, Google, and Microsoft can integrate with Nvidia GPUs
DRAM becomes a hot commodity as inference needs increase