single mention · inference & serving
Latency-sensitive execution class
Separates latency-sensitive workloads into an execution class; the evidence also describes priority isolation for that class.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See inference scheduling for the methods filed beside it.
Also called a latency-sensitive (LS) execution class.
Filed under inference & serving › inference scheduling