METHOD AND APPARATUS FOR FAULTY MEMORY UTILIZATION

• US 20150074367A1
• Filed: 09/09/2013
• Published: 03/12/2015
• Est. Priority Date: 09/09/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Cher, Chen-Yong, Park, Yoonho, Rosenburg, Bryan S., Ryu, Kyung D., Andrade Costa, Carlos H.
• Granted Patent: US 9,317,350 B2
• Time in Patent Office: Days
• US Class Current: 711/165
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/07 Responding to the occurrence of a fault, e.g. fault tolerance
• CPC Class Codes: G06F 11/073 in a memory management context, e.g. virtual memory or cache management memory management G06F12/00; testing of static memory units G11C29/00
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit
• CPC Class Codes: G06F 11/3037 where the computing system component is a memory, e.g. virtual memory, cache accessing, addressing or allocating within memory systems or architectures G06F12/00; checking stores for correct operation G11C29/00
• CPC Class Codes: G06F 11/3058 Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations thermal management in cooling arrangements of a computing system G06F1/206
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/88 Monitoring involving counting

Abstract

A method for faulty memory utilization in a memory system includes: obtaining information regarding memory health status of at least one memory page in the memory system; determining an error tolerance of the memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory system affecting the memory page; initiating a migration of data stored in the memory page when it is determined that the data stored in the memory page is non-error-tolerant; notifying at least one application regarding a predicted operating system failure and/or a predicted application failure when it is determined that data stored in the memory page is non-error-tolerant and cannot be migrated; and notifying at least one application regarding the memory failure predicted to occur when it is determined that data stored in the memory page is error-tolerant.

Claims

1. A method for facilitating faulty memory utilization in a memory system, the method comprising:

obtaining information regarding memory health status of at least one memory page in the memory system;

determining an error tolerance of the at least one memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory system affecting the at least one memory page;

initiating a migration of data stored in the at least one memory page when it is determined that the data stored in the at least one memory page is non-error-tolerant;

notifying at least one application regarding at least one of a predicted operating system failure and a predicted application failure when it is determined that data stored in the at least one memory page is non-error-tolerant and cannot be migrated;

notifying at least one application regarding the memory failure predicted to occur when it is determined that data stored in the at least one memory page is error-tolerant, whereby the at least one memory page can continued to be used.

2. The method of claim 1, wherein the step of initiating a migration of data comprises migrating the data stored in the at least one memory page to a memory page determined to be unaffected by a predicted memory failure in the memory system.

3. The method of claim 1, wherein the step of determining the error tolerance of the at least one memory page comprises assessing an error tolerance of data in at least one corresponding virtual memory page to be affected by a predicted memory failure in the memory system.

4. The method of claim 3, wherein the step of assessing the error tolerance of data in the at least one corresponding virtual memory page comprises:

identifying at least one memory page mapped to corresponding physical memory to be affected by the predicted memory failure; and

classifying the data into one of error-tolerant data and non-error-tolerant data.

5. The method of claim 4, wherein migratable kernel-space pages are classified as non-error-tolerant data and user-space pages are classified as error-tolerant data or non-error-tolerant data.

6. The method of claim 1, wherein the step of obtaining information regarding memory health status of at least one memory page comprises receiving a notification indicating at least one of a memory range to be affected by a predicted memory failure, an expected time before the memory failure is likely to occur, and an uncorrectable memory error rate expected after occurrence of the memory failure.

7. The method of claim 1, wherein the step of initiating a migration of data comprises:

locating at least one memory page unaffected by a predicted memory failure;

obtaining a time before the predicted memory failure is likely to occur;

determining whether there is sufficient time to migrate the data based on at least one of the time before the predicted memory failure is likely to occur and an anticipated time to complete the migration of data.

8. The method of claim 1, further comprising:

obtaining an error tolerance threshold when it is determined that the data stored in the at least one memory page is error-tolerant;

determining whether the error tolerance threshold has been exceeded;

sending a notification to an application indicating an expected uncorrectable error rate corresponding to the at least one memory page when it is determined that the data stored in the at least one memory page is error-tolerant;

sending a memory address range of the at least one memory page to the application when it is determined that the data stored in the at least one memory page is error-tolerant, whereby the at least one memory page continues to be used by the application.

9. The method of claim 8, wherein the step of sending a notification to an application comprises generating a notification signal indicating an area of the memory system predicted to fail and a time period before predicted memory failure, the notification signal being used in application-level fault management.

10. The method of claim 1, wherein the method is performed without disrupting execution of an application running on the memory system.

11. The method of claim 1, further comprising:

monitoring an uncorrected error rate affecting error-tolerant data allocated in at least one memory area including a predicted memory failure; and

ensuring that the uncorrected error rate is below an error-tolerance threshold defined in an application running on the memory system.

12. The method of claim 1, further comprising providing an application programming interface, wherein the step of determining an error tolerance of the at least one memory page comprises identifying at least one data region adapted to tolerate a prescribed degree of data corruption via the application programming interface.

13. The method of claim 12, further comprising enabling a programmer to identify data regions that are error-tolerant when writing an application utilizing the application programming interface.

14. An apparatus for facilitating faulty memory utilization in a memory system, the apparatus comprising:

a memory;

a health tracking module operative to obtain information regarding memory health status of at least one memory page in the memory; and

at least one processor in communication with the memory and health tracking module, the at least one processor being operative;

to receive the information regarding memory health status;

to determine an error tolerance of the at least one memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory affecting the at least one memory page;

to initiate a migration of data stored in the at least one memory page when it is determined that the data stored in the at least one memory page is non-error-tolerant;

to notify at least one application regarding at least one of a predicted operating system failure and a predicted application failure when it is determined that data stored in the at least one memory page is non-error-tolerant and cannot be migrated; and

to notify at least one application regarding the memory failure predicted to occur when it is determined that data stored in the at least one memory page is error-tolerant, whereby the at least one memory page can continued to be used.

15. The apparatus of claim 14, wherein the at least one processor comprises a memory manager coupled with the memory and operative to control a storage of at least one of data and programs in the memory, the memory manager being configured to receive the information regarding memory health status from the health tracking module.

16. The apparatus of claim 14, wherein the at least one processor comprises a memory page assessment module operative to receive information regarding a type of data stored in one or more memory pages of the memory and to determine an error tolerance of at least one of the memory pages as a function of the type of data stored in the memory pages when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory affecting the at least one memory page.

17. The apparatus of claim 14, wherein the at least one processor comprises a memory page reassignment module operative to receive a notification signal indicating a predicted imminent failure to a non-error-tolerant memory area and to initiate a proactive memory page migration action in response to the notification signal when it is determined that the data stored in the at least one memory page is non-error-tolerant.

18. The apparatus of claim 17, wherein the memory page reassignment module is further operative to notify at least one application regarding at least one of a predicted operating system failure and a predicted application failure when it is determined that data stored in the at least one memory page is non-error-tolerant and cannot be migrated, and to notify at least one application regarding the memory failure predicted to occur when it is determined that data stored in the at least one memory page is error-tolerant.

19. The apparatus of claim 14, further comprising an application program interface operative to determine an error tolerance of the at least one memory page by identifying at least one data region adapted to tolerate a prescribed degree of data corruption.

20. A computer program product for facilitating faulty memory utilization in a memory system, comprising a computer readable storage medium having computer readable program code embodied therewith, said computer readable program code comprising:

computer readable program code configured to obtain information regarding memory health status of at least one memory page in a memory system;

computer readable program code configured to determine an error tolerance of the at least one memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory system affecting the at least one memory page;

computer readable program code configured to initiate a migration of data stored in the at least one memory page when it is determined that the data stored in the at least one memory page is non-error-tolerant;

computer readable program code configured to notify at least one application regarding at least one of a predicted operating system failure and a predicted application failure when it is determined that data stored in the at least one memory page is non-error-tolerant and cannot be migrated; and

computer readable program code configured to notify at least one application regarding the memory failure predicted to occur when it is determined that data stored in the at least one memory page is error-tolerant, whereby the at least one memory page can continued to be used.