Industry Insights
Find the latest news, analysis and insight on enterprise IT infrastructure, data centre hardware, AI systems and the technology trends that matter. Written for IT professionals and business buyers who want to stay informed without the fluff.
MLPerf Inference v6.1: what 486 results tell buyers about inference cost per rack
MLPerf Inference v6.1 brought a record 30 submitters and 486 datacentre and edge results, the first peer reviewed Vera Rubin...
HPE Alletra MP X10000 release 4 doubles cluster size to 16 nodes and adds NFS
HPE Alletra Storage MP X10000 release 4 is generally available, lifting the supported limit from 8 nodes and JBOFs to...
AWS Puts Outposts In One 42U Rack: What A 2,688 vCPU On-Premise Cloud Means For Space Constrained Sites
AWS has made second-generation single-rack Outposts generally available: compute, storage and networking in one 42U cabinet with up to 2,688...
The AI buildout's decommissioning bill: why power, cooling and network gear dominate the waste, not GPUs
A new Basel Action Network report models 395 to 617 million tonnes of AI-related electronic waste between 2025 and 2050,...
AMD's EPYC Venice white paper puts numbers on Zen 6, but read the compiler footnotes
AMD has published its first detailed benchmark white paper for EPYC Venice, claiming the 256-core flagship is 2.24x faster than...
Qualcomm and AWS Commit to Multi-Generation Inference Silicon: What Custom Accelerators Mean for Everyone Else's Buying Plans
Qualcomm and Amazon have agreed a multi-generation collaboration on custom AI inference silicon and optical connectivity up to 1.6T for...
Energised Power, Not GPUs, Is Now The Limiting Factor In AI Cloud Buildouts
AI cloud provider Corvex has raised 33 million dollars to grow from 1.5MW to 8MW by the end of 2026,...
ASUS moves from AI servers to full AI factory platforms: what changes for buyers
At AI Tech 2026 in Seoul, ASUS set out an AI factory platform spanning rack-scale Vera Rubin systems, AI-native and...
