Brief inference latency spike (us-east-1)
Operationalstarted APR 15 · 07:12:00resolved APR 15 · 07:16:00
modelsingest
- operationalAPR 15 · 07:16:00 · 1d agoAuto-scaler restored capacity. Latency is back to baseline.
- investigatingAPR 15 · 07:12:00 · 1d agop95 inference latency jumped to 320 ms in us-east-1.