The report of what the accelerator performance model predicts for the target workloads, with the assumptions, sensitivities and recommendations that set the performance targets.
Report Work product in the AI accelerator lifecycle
It turns modelling into evidence that product planning, numerics and architecture decisions can rely on. Targets set without a stated basis leave the architecture team unable to show why parameters were chosen and the product plan promising numbers nobody has checked.
The full lifecycle detail, with its charts, flows, tables and every section, is for registered members. Registration is free and takes a minute. You come back to this page afterwards.