Optimized recovery in data replication environments

• US 9,946,617 B2
• Filed: 06/06/2016
• Issued: 04/17/2018
• Est. Priority Date: 06/06/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Crawford, Joshua J., Jennas, II, Paul A., Peipelman, Jason L., Ward, Matthew J.
• Primary Examiner(s): MASKULINSKI, MICHAEL C
• Application Number:
  US15/174,742
• Publication Number:
  US 20170351591A1
• Time in Patent Office: 680 Days
• CPC Class Codes: G06F 11/07 Responding to the occurrence of a fault, e.g. fault tolerance
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/2064 while ensuring consistency
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 11/2082 Data synchronisation
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/3034 where the computing system component is a storage system, e.g. DASD based or network based digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/81 Threshold

Abstract

A method for optimizing recovery in a data replication environment is disclosed. In one embodiment, such a method includes directing I/O from a primary site to a secondary site in response to a failure at the primary site. After the primary site has recovered from the failure, the method initiates a recovery process wherein updated data elements at the secondary site are copied to the primary site. The method determines a recorded average I/O latency for a host system driving I/O to the secondary site, and calculates an expected average I/O latency for the host system driving I/O to the primary site. The method redirects I/O from the secondary site to the primary site when a difference between the expected average I/O latency and the recorded average I/O latency reaches a threshold value. A corresponding system and computer program product are also disclosed.

Claims

1. A method for optimizing recovery in a data replication environment, the method comprising:

directing I/O from a primary site to a secondary site in response to a failure at the primary site;

after the primary site has recovered from the failure, initiating a recovery process wherein updated data elements at the secondary site are copied to the primary site;

determining a recorded average I/O latency for a host system driving I/O to the secondary site;

calculating an expected average I/O latency for the host system driving I/O to the primary site; and

redirecting I/O from the secondary site to the primary site when a difference between the expected average I/O latency and the recorded average I/O latency reaches a threshold value.

2. The method of claim 1, wherein the threshold value is zero.

3. The method of claim 1, further comprising determining a first probability that a read to the primary site will require retrieving data from the secondary site, and a second probability that a read to the primary site will not require retrieving data from the secondary site.

4. The method of claim 3, further comprising calculating a first latency component by multiplying the first probability by the sum of:

(1) an average I/O latency between the host system and the primary site; and

(2) an average I/O latency between the primary site and the secondary site.

5. The method of claim 4, further comprising calculating a second latency component by multiplying the second probability by the average I/O latency from the host system to the primary site.

6. The method of claim 5, wherein calculating the expected average I/O latency comprises adding the first latency component to the second latency component.

7. The method of claim 1, wherein calculating the expected average I/O latency comprises repeatedly recalculating the expected average I/O latency during the recovery process.

8. A computer program product for optimizing recovery in a data replication environment, the computer program product comprising a computer-readable medium having computer-usable program code embodied therein, the computer-usable program code comprising:

computer-usable program code to direct I/O from a primary site to a secondary site in response to a failure at the primary site;

computer-usable program code to, after the primary site has recovered from the failure, initiate a recovery process wherein updated data elements at the secondary site are copied to the primary site;

computer-usable program code to calculate a recorded average I/O latency for a host system driving I/O to the secondary site;

computer-usable program code to calculate an expected average I/O latency for the host system driving I/O to the primary site; and

computer-usable program code to redirect I/O from the secondary site to the primary site when a difference between the expected average I/O latency and the recorded average I/O latency reaches a threshold value.

9. The computer program product of claim 8, wherein the threshold value is zero.

10. The computer program product of claim 8, further comprising computer-usable program code to determine a first probability that a read to the primary site will require retrieving data from the secondary site, and a second probability that a read to the primary site will not require retrieving data from the secondary site.

11. The computer program product of claim 10, further comprising computer-usable program code to calculate a first latency component by multiplying the first probability by the sum of:

(1) an average I/O latency between the host system and the primary site; and

(2) an average I/O latency between the primary site and the secondary site.

12. The computer program product of claim 11, further comprising computer-usable program code to calculate a second latency component by multiplying the second probability by the average I/O latency from the host system to the primary site.

13. The computer program product of claim 12, wherein calculating the expected average I/O latency comprises adding the first latency component to the second latency component.

14. The computer program product of claim 8, wherein calculating the expected average I/O latency comprises repeatedly recalculating the expected average I/O latency during the recovery process.

15. A system for optimizing recovery in a data replication environment, the system comprising:

at least one processor;

at least one memory device coupled to the at least one processor and storing instructions for execution on the at least one processor, the instructions causing the at least one processor to;

direct I/O from a primary site to a secondary site in response to a failure at the primary site;

after the primary site has recovered from the failure, initiate a recovery process wherein updated data elements at the secondary site are copied to the primary site;

calculate a recorded average I/O latency for a host system driving I/O to the secondary site;

calculate an expected average I/O latency for the host system driving I/O to the primary site; and

redirect I/O from the secondary site to the primary site when a difference between the expected average I/O latency and the recorded average I/O latency reaches a threshold value.

16. The system of claim 15, wherein the threshold value is zero.

17. The system of claim 15, wherein the instructions further cause the at least one processor to determine a first probability that a read to the primary site will require retrieving data from the secondary site, and a second probability that a read to the primary site will not require retrieving data from the secondary site.

18. The system of claim 17, wherein the instructions further cause the at least one processor to calculate a first latency component by multiplying the first probability by the sum of:

(1) an average I/O latency between the host system and the primary site; and

(2) an average I/O latency between the primary site and the secondary site.

19. The system of claim 18, wherein the instructions further cause the at least one processor to calculate a second latency component by multiplying the second probability by the average I/O latency from the host system to the primary site.

20. The system of claim 19, wherein calculating the expected average I/O latency comprises adding the first latency component to the second latency component.