Methods, systems and devices for recovering from corruptions in data processing units

A method of responding to a process failure of a data processing unit (DPU) of a plurality of DPUs of a computing system without a central processing unit and each DPU communicatively coupled to a non-volatile memory, the method at a first DPU comprising:

• receiving instructions to execute a parent process from a second data processing unit;
• transmitting instructions to execute a child process associated with the parent process to a third data processing unit;
• receiving status information pertaining to the child process indicating a process failure at the third data processing unit;
• and
• re-assigning the child process for execution.

An apparatus comprising:

• means for receiving instructions to execute a parent process at a first data processing unit, from a second data processing unit of a plurality of data processing units of a computing system without a central processing unit and each data processing unit communicatively coupled with a non-volatile memory;
• means for transmitting instructions to execute a child process associated with the parent process to a third data processing unit;
• means for receiving status information pertaining to the child process indicating a process failure at the third data processing unit;
• and
• means for re-assigning the child process for execution.

A computing system comprising a device, the device comprising:

• a non-volatile memory divided into a plurality of memory sub-arrays, each memory sub-array comprising a plurality of selectable locations;
• and
• a plurality of data processing units communicatively coupled to the non-volatile memory in the absence of a central processing unit of the computing system, including a first data processing unit assigned to process data of a memory sub-array, and configured to:
• receive instructions to execute a parent process from a second data processing unit;
• transmit instructions to execute a child process associated with the parent process to a third data processing unit;
• detect occurrence of a processing issue at the first data processing unit;
• parse a jobs queue at the first data processing unit for unexecuted processes in response to detecting the occurrence of the processing issue at the first data processing unit;
• parse a tasks queue at the first data processing unit for unexecuted processes in response to detecting the occurrence of the processing issue at the first data processing unit;
• and
• re-assign the child process for execution.