US 11,727,309 B2
Runtime estimation for machine learning tasks
Parijat Dube, Yorktown Heights, NY (US); Gauri Joshi, Pittsburgh, PA (US); Priya Ashok Nagpurkar, Ossining, NY (US); Stefania Costache, White Plains, NY (US); Diana Jeanne Arroyo, Austin, TX (US); and Zehra Noman Sura, Yorktown Heights, NY (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Oct. 28, 2021, as Appl. No. 17/452,596.
Application 17/452,596 is a continuation of application No. 15/901,430, filed on Feb. 21, 2018, granted, now 11,200,512.
Prior Publication US 2022/0051142 A1, Feb. 17, 2022
Int. Cl. G06N 20/00 (2019.01); G06F 16/22 (2019.01)
CPC G06N 20/00 (2019.01) [G06F 16/22 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory that stores computer executable components;
a processor, operably coupled to the memory, and that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a monitoring component that monitors one or more operating conditions of a cloud computing environment while one or more machine learning tasks are executed;
a model component that generates an estimation model for estimating runtimes of machine learning tasks, based on the one or more operating conditions, respective runtimes times for completed machine learning tasks of the one or more machine learning tasks that were executed, and respective machine learning model types and respective frameworks of the completed machine learning tasks; and
an estimation component that generates an estimated runtime of a machine learning task based on the estimation model, a machine learning model type of the machine learning task, and at least one current operating condition of the cloud computing environment, wherein the estimated runtime defines a period of time beginning at an initiation of the machine learning task and ending at a completion of the machine learning task, and the at least one completed machine learning task does not comprise the machine learning task.