Automatic correcting of computing cluster execution failure

Patent metadata

Field	Value
Publication number	US-2019317849-A1
Application number	US-201815950260-A
Country	US
Kind code	A1
Filing date	Apr 11, 2018
Priority date	Apr 11, 2018
Publication date	Oct 17, 2019
Grant date	—

Abstract

A processor may identify, using historical data, an amount of computing resources consumed to remedy the failure with an automatic remedy step. The processor may determine that the amount of consumed computing resources to remedy the failure is less than an amount of computing resources consumed by restarting the process. The processor may perform the automatic remedy step. The processor may identify that the automatic remedy step has failed. The processor may determine a waiting period based on an estimated time to receive a user response to the failure and an estimated load on the computing cluster. The processor may display a generated alert to a user during the waiting period. The processor may identify that no user input has been received during the waiting period. The processor may release computing resources corresponding to the process.

First claim

1 - 12 . (canceled)

13 . A computer program product comprising a non-transitory computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method, the method comprising:
identifying, using historical resource data, an amount of computing resources consumed to remedy the failure with an automatic remedy step; determining that the amount of consumed computing resources to remedy the failure is less than an amount of computing resources consumed by restarting the process; performing the automatic remedy step; identifying that the automatic remedy step has failed; determining a waiting period based on an estimated time to receive a user response to the failure and an estimated load on the computing cluster; displaying a generated alert to a user during the waiting period, wherein the alert includes a prompt for user input regarding the user response to the failure; identifying that no user input has been received during the waiting period; and releasing computing resources corresponding to the process.

14 . The computer program product of claim 13 , wherein the process includes multiple execution stages, and wherein the method further comprises:
saving results of already executed stages of the process before permitting the release of computing resources; restarting the process; and stopping the process based on the execution stage of the process at which the failure occurred.

15 . The computer program product of claim 13 , further comprising:
determining a waiting cost based on the estimated time to receive the user response to the failure and the estimated load on the computing cluster, wherein the waiting cost is the amount of computing resources consumed during the waiting period; identifying that the waiting cost is lower than the amount of computing resources consumed by restarting the process; and issuing the alert.

16 . A system comprising:
a memory; and a processor in communication with the memory, the processor being configured to perform operations comprising: identifying, using historical resource data, an amount of computing resources consumed to remedy the failure with an automatic remedy step; determining that the amount of consumed computing resources to remedy the failure is less than an amount of computing resources consumed by restarting the process; performing the automatic remedy step; identifying that the automatic remedy step has failed; determining a waiting period based on an estimated time to receive a user response to the failure and an estimated load on the computing cluster; displaying a generated alert to a user during the waiting period, wherein the alert includes a prompt for user input regarding the user response to the failure; identifying that no user input has been received during the waiting period; and releasing computing resources corresponding to the process.

17 . The system of claim 16 , wherein the process includes multiple execution stages, and wherein the method further comprises:
saving results of already executed stages of the process before permitting the release of computing resources; restarting the process; and stopping the process based on the execution stage of the process at which the failure occurred.

18 . The system of claim 16 , further comprising:
determining a waiting cost based on the estimated time to receive the user response to the failure and the estimated load on the computing cluster, wherein the waiting cost is the amount of computing resources consumed during the waiting period; identifying that the waiting cost is lower than the amount of computing resources consumed by restarting the process; and issuing the alert.

19 . The system of claim 16 , wherein the process includes multiple execution stages, wherein the historical resource data comprises an amount of consumed computing resources associated with each execution stage, and wherein the amount of consumed computing resources to remedy the failure is determined using the amount of consumed computing resources associated with the stage of execution of the process at which the failure occurred.

20 . The system of claim 16 , further comprising:
generating a timer, wherein the timer maintains a first function of the process during the waiting period; generating one or more substitute tasks for the first function; and executing the one or more substitute tasks during the waiting period.

Assignees

• IBM

Inventors

• Cmielowski Lukasz G
• Slowikowski Pawel
• Bigaj Rafal
• Malecki Bartlomiej

Classifications

• G06F11/0772

  Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers · CPC title

• G06F11/079

  Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36) · CPC title

• G06F11/0793 Primary

  Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40) · CPC title

• G06F11/0751

  Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06) · CPC title

• G06F11/0721

  within a central processing unit [CPU] · CPC title