Systems and methods for autonomous hardware compute resiliency

Abstract

Methods and systems for providing hardware compute resiliency by using a compute fabric that includes sensors and re-programmable data processing components.

What is claimed is: 
     
         1 . A method for controlling a system that includes a data collection system, at least one sensor, and a hardware compute fabric, the method comprising:
 the data collection system generating data;   a first processing component of the compute fabric performing a first sub-process of a data analysis process for the generated data and storing results of the first sub-process at the compute fabric;   a third processing component programming a second processing component of the compute fabric to perform the first sub-process and disabling the first sub-process at the first processing component, based on at least one of environment information generated by using the at least one sensor, power consumption information generated by the compute fabric, processing performance information for the data analysis process generated by the compute fabric, and fault detection information for the data analysis process generated by the compute fabric;   the second processing component performing the first sub-process for the generated data and storing results of the first sub-process generated by the second processing component at the compute fabric;   the compute fabric generating data analytics results by using the stored results of the first sub-process generated by the first processing component and the stored results of the first sub-process generated by the second processing component; and   a data transmitter system of the compute fabric transmitting the generated data analytics results to at least one external system.   
     
     
         2 . A system comprising:
 a data collection system constructed to generate data;   at least one sensor;   a data transmitter system;   a first processing component that is constructed to perform a first sub-process of a data analysis process for the generated data and store results of the first sub-process;   a third processing component that is constructed to program a second processing component to perform the first sub-process and disable the first sub-process at the first processing component, based on at least one of environment information generated by using the at least one sensor, power consumption information, processing performance information for the data analysis process, and fault detection information for the data analysis process; and   the second processing component, wherein the second processing component is constructed to perform the first sub-process for the generated data and store results of the first sub-process generated by the second processing component,   wherein the system is constructed to generate data analytics results by using the stored results of the first sub-process generated by the first processing component and the stored results of the first sub-process generated by the second processing component, and   wherein the data transmitter system is constructed to transmit the generated data analytics results to at least one external system.