公开(公告)号：US10853129B1

公开(公告)日：2020-12-01

申请号：US16358355

申请日：2019-03-19

Applicant: Amazon Technologies, Inc.

Inventor： Sudipta Sengupta ,  Haifeng He ,  Pejus Manoj Das ,  Poorna Chand Srinivas Perumalla ,  Wei Xiao ,  Shirley Xue Yi Leung ,  Vladimir Mitrovic ,  Yongcong Luo ,  Jiacheng Guo ,  Stefano Stefani ,  Matthew Shawn Wilson

IPC: G06F9/46 ,  G06F9/48 ,  G06N20/00 ,  G06N5/04 ,  G06F9/50 ,  G06N3/08 ,  G06F9/455

Abstract: Implementations detailed herein include description of a computer-implemented method to migrate a machine learning model from one accelerator portion (such as a portion of a graphical processor unit (GPU)) to a different accelerator portion. In some instances, a state of the first accelerator portion is persisted, the second accelerator portion is configured, the first accelerator portion is then detached from a client application instance, and at least a portion of an inference request is performed using the loaded at least a portion of the machine learning model on the second accelerator portion that had been configured.