{"event":{"id":"evt-tpuv4-hub-20220511","dedupe_key":"public_ai_compute_infrastructure:2022-05-11:google-cloud-puts-an-eight-pod-tpu-v4-ml-hub-into-public-preview","event_type":"public_ai_compute_infrastructure","title":"Google Cloud puts an eight-pod TPU v4 ML hub into public preview","summary":"Google Cloud announced public preview of a TPU v4 ML hub with eight pods, 4,096 chips per pod, high-bandwidth interconnect and nearly nine exaflops aggregate peak compute, marketed with a 90% carbon-free energy profile.","occurred_at":"2022-05-11T00:00:00.000Z","published_at":"2022-05-11T00:00:00.000Z","observed_at":"2026-10-01T18:14:00.000Z","reconstructed_at":"2026-10-01T18:14:00.000Z","ingest_type":"live_world_scan","locations":null,"status":"verified","confidence":0.99,"metadata":{"actors":["Google Cloud"],"what_happened_at_time":"Very large accelerator clusters that had primarily characterized frontier internal infrastructure became externally accessible as cloud infrastructure in preview.","evidence_at_time":"Google Cloud documented pod size, network bandwidth, price-performance goals, public preview status and the Oklahoma site's carbon-free energy profile.","affected_layers":["layer-compute","layer-chips","layer-networks","layer-software","layer-research","layer-models","layer-energy"],"change_kind":"compute_access_maturation","q1_continuity":"Q1 established dedicated AI supercomputing as strategic. Q2 widened access and made price-performance, interconnect and infrastructure software more explicit.","retrospective_inference":"Compute was becoming a service layer that could expose frontier-scale hardware to external researchers and developers rather than remaining only an internal lab asset.","uncertainty":"Preview availability did not mean unconstrained supply; the source itself framed scale and price-performance as pain points. Energy was explicit as a sustainability/siting attribute, not demonstrated as a binding AI power bottleneck."},"created_at":"2026-10-01T19:22:17.747Z","updated_at":"2026-10-01T19:22:17.747Z"},"artifacts":[{"id":"art-tpuv4-hub-20220511","canonical_url":"https://cloud.google.com/blog/products/compute/google-unveils-worlds-largest-publicly-available-ml-cluster","artifact_type":"cloud_compute_preview","title":"Cloud TPU v4 ML hub public preview","summary":"Google Cloud put an eight-pod TPU v4 ML hub into public preview, with 4,096 chips per pod, high-bandwidth interconnect, nearly nine exaflops aggregate peak compute and a stated 90% carbon-free energy profile.","creator_entities":["Google Cloud"],"released_at":"2022-05-11T00:00:00.000Z","content_hash":null,"metadata":{"historical_scope":"2022-Q2"},"created_at":"2026-10-01T19:22:17.741Z","updated_at":"2026-10-01T19:22:17.741Z","relation":"documented_by"}],"signals":[{"id":"sig-q2-energy-carbon-metric","statement":"Energy and carbon intensity became explicit metrics in AI infrastructure and large-model training announcements, but Q2 evidence does not establish electricity supply as a binding AI compute bottleneck.","signal_type":"infrastructure_metric_visibility","direction":"increasing","confidence":0.83,"epistemic_status":"supported_inference","mapping_mode":"retrospective","reconstructed_at":"2026-10-01T18:14:00.000Z","created_at":"2026-10-01T19:22:17.756Z","updated_at":"2026-10-01T19:22:17.756Z"},{"id":"sig-q2-public-frontier-compute","statement":"Frontier-scale accelerator infrastructure became more externally accessible through cloud preview, while scale, price-performance, interconnect and software utilization remained explicit constraints.","signal_type":"compute_access_maturation","direction":"increasing","confidence":0.97,"epistemic_status":"supported_inference","mapping_mode":"retrospective","reconstructed_at":"2026-10-01T18:14:00.000Z","created_at":"2026-10-01T19:22:17.754Z","updated_at":"2026-10-01T19:22:17.754Z"},{"id":"sig-q2-scale-orchestration","statement":"Frontier model scaling increasingly depended on distributed-systems software and high-bandwidth multi-pod networking, not only on adding accelerator chips.","signal_type":"stack_coupling_strengthening","direction":"increasing","confidence":0.97,"epistemic_status":"supported_inference","mapping_mode":"retrospective","reconstructed_at":"2026-10-01T18:14:00.000Z","created_at":"2026-10-01T19:22:17.752Z","updated_at":"2026-10-01T19:22:17.752Z"}]}