International Business Machines Corporation
SELF-RELOCATING DATA CENTER BASED ON PREDICTED EVENTS

Last updated:

Abstract:

A self-relocating data center and method is provided. In implementations, a method includes: receiving input data from a data center over a period of time, the input data including sensor data; determining parameter data sets based on the input data, the parameter data sets reflecting operating states of resources of the data center over the period of time; determining dependencies between the parameter data sets using a stacked long short term memory (LSTM) process; generating a model based on the dependencies between the respective parameter data sets and stored target data parameters; determining an event based on an output of the model and the stored target parameter data, wherein an input to the model is real-time input data received from the data center; and initiating a self-relocating event to automatically relocate data of the data center to a secondary data center without human intervention based on the event.

Status:
Application
Type:

Utility

Filling date:

8 May 2020

Issue date:

11 Nov 2021