ARCHITECTURE
TPU v1 accelerator architecture
First-generation inference architecture with systolic matrix arithmetic, explicit buffers, and host-directed operations.
TPU v1 architecture
Specifications
0 SOURCE-LINKED FIELDSSpecifications have not yet been established for this record.
THE VISUAL EVIDENCE
0 DOCUMENTED SOURCESVisual sources
Help document this silicon.
We have not yet identified a reliable public visual source for this record. Suggest a die photograph, package image, paper or manufacturer diagram.
Follow the connections
2 SOURCE-LINKED RELATIONSHIPSOrganization identified in the cited documentation; this edge does not imply foundry or packaging responsibility.
Norman P. Jouppi et al. / Google; ISCA 2017 ยท EvidenceSources and historical context
In-Datacenter Performance Analysis of a Tensor Processing Unit
Norman P. Jouppi et al. / Google; ISCA 2017 ยท official-disclosure ยท publicly-readable
An in-depth look at Googleโs first Tensor Processing Unit (TPU)
Google Cloud ยท official-disclosure ยท publicly-readable