compute.layer.re · historical view at 2026-09-30
AI compute expansion continues through integrated rack-scale deployments, but memory capacity and grid-cost allocation remain simultaneous constraints.
Micron reported AI-driven HBM orders above current capacity, broader memory-supply tightening and continued investment expansion to meet demand.
evt-micron-hbm-demand-20260930The Senate voted 57-43 to advance legislation directing consideration of cost-allocation standards for large electricity users such as data centers, below the 60 votes needed.
evt-us-senate-dc-cost-vote-20260930HPE announced a $1.2 billion order for AMD Helios AI racks at Vultr US data centers, integrating accelerators, networking, software and liquid cooling.
evt-hpe-vultr-helios-order-20260930DeepSeek said it was open-sourcing programming infrastructure for Huawei Ascend chips and jointly advancing supernode software, including TileLang-related work.
evt-deepseek-huawei-stack-20260930NVIDIA released OpenShell and the Sentry reference design to enforce agent boundaries across software and hardware, including policy, identity, tool and data-access controls.
evt-nvidia-agent-safety-20260928Anthropic and Akamai entered a seven-year $11.6 billion cloud services agreement, with Akamai planning major supporting capital expenditure and component purchases.
evt-anthropic-akamai-cloud-20260924The European Commission proposed a common data-center rating scheme and opened work toward minimum performance standards covering energy and water use, efficiency and grid integration.
evt-eu-datacenter-framework-20260921Huawei announced a large-scale AI compute system built around UnifiedBus, NPO interconnects and up to 4,096 NPUs, alongside continued CANN open-source development.
evt-huawei-atlas960e-20260917DeepMind's March 29 preprint argued that many large language models were undertrained for their compute budgets and reported better performance by balancing model size with substantially more training tokens.
evt-chinchilla-20220329NVIDIA announced H100 with a Transformer Engine and NVLink scaling plus DGX H100 systems and SuperPOD architecture for large language models and other large AI workloads.
evt-nvidia-h100-20220322CoreWeave and EleutherAI announced GPT-NeoX-20B public API access and an imminent full open-source release, expanding the open-model path for large language model research and applications.
evt-neox-public-20220202Meta announced RSC as operational with 6,080 A100 GPUs connected through high-performance networking and large storage, explicitly tying larger model research to dedicated supercomputing infrastructure.
evt-meta-rsc-20220124