Method for reliably updating a data group in a read-before-write data replication environment using a comparison file

• US 8,572,331 B2
• Filed: 10/30/2008
• Issued: 10/29/2013
• Est. Priority Date: 06/14/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Butterworth, Henry Esmond, Wolfgang, John Jay, Zautner, Vitaly, Zlotnick, Aviad, Day, Kenneth Fairclough III, Doatmas, Philip Matthew
• Primary Examiner(s): Shah, Sanjiv
• Assistant Examiner(s): Ayash, Marwan
• Application Number:
  US12/261,972
• Publication Number:
  US 20090043826A1
• Time in Patent Office: 1,825 Days
• Field of Search: None
• US Class Current: 711/154
• CPC Class Codes: G06F 11/08 Error detection or correction by redundancy in data representation, e.g. by using checking codes
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

A method is disclosed for reliably updating a data group in a read-before-write data replication environment. The method reliably updates the data group by receiving an updated data group sent from a first storage medium to a second storage medium, comparing the updated data group with a previous data group previously existing on the second storage medium and writing the updated data group to the second storage medium. The read-before-write and differencing method disclosed maintain reliability by storing multiple copies of changes made to the second storage medium during and after the write process.

Claims

1. A program of executable code stored on a non-transitory storage device and executed by a processor to perform operations to reliably update a data group in a read-before-write replication environment, the operations comprising:

receiving at a storage controller separate from a first non-volatile storage medium an updated data group sent from the first non-volatile storage medium in a read-before-write operation, wherein the updated data group comprises unchanged data values and updated data values to replace previous data values;

copying a previous data group stored on a second non-volatile storage medium separate from the first non-volatile storage medium into a controller memory of the storage controller;

comparing the updated data group with the previous data group;

wherein the updated data group is compared with the previous data group by applying an Exclusive OR (XOR) command;

storing results of the comparison of the updated data group and the previous data group in a comparison file, the comparison file comprising the differences between updated data values in the updated data group and the previous data values in the previous data group;

compressing the comparison file;

communicating the comparison file and the updated data group to a peer controller;

initiating a write process wherein the storage controller writes the updated data group to a t-he second non-volatile storage medium; and

detecting a failure of the write process of the updated data group to the second nonvolatile storage medium, wherein the peer controller is configured to complete the read-before- write operation in place of the storage controller.

2. The program of claim 1, the operations further comprising informing the peer controller prior to writing the updated data group to the second non-volatile storage medium.

3. The program of claim 1, the operations further comprising informing the peer controller that the updated data group is written to the second non-volatile storage medium.

4. The program of claim 3, the operations further comprising discarding the updated data group from the peer controller when the peer controller is informed that the updated data group was written to the second non-volatile storage medium.

5. The program of claim 1, the operations further comprising:

reading the updated data group from the second non-volatile storage medium and the comparison file into the controller memory of the storage controller;

recovering the previous data group from the updated data group and the comparison file; and

writing the recovered previous data group to the second non-volatile storage medium.

6. The program of claim 5, wherein the previous data group is recovered from the updated data group and the comparison file by applying the Exclusive OR (XOR) command to the updated data group and the comparison file.

7. The program of claim 5, the operations further comprising:

determining to roll forward the updated data group;

recovering the updated data group from the previous data group and the comparison file; and

writing the recovered updated data group to the second non-volatile storage medium.

8. A method for reliably updating a data group in a read-before-write replication environment, the method performed by executable code stored on a non-transitory storage device and executed by a processor and comprising:

receiving at a storage controller separate from a first non-volatile storage medium an updated data group sent from the first nonvolatile storage medium in a read-before-write operation, wherein the updated data group comprises unchanged data values and updated data values to replace previous data values;

copying a previous data group stored on a second non-volatile storage medium into a controller memory of the storage controller;

comparing the updated data group with the previous data group using an Exclusive OR (XOR) command;

storing results of the comparison of the updated data group and the previous data group as a comparison file, the comparison file comprising the differences between updated data values in the updated data group and the previous data values in the previous data group;

compressing the comparison file;

informing a peer controller prior to writing the updated data group to the second nonvolatile storage medium, the peer controller configured to complete the read-before write operation for the storage controller in case of failure;

communicating the comparison file and the updated data group to the peer controller;

synchronizing the updated data group to the peer controller;

writing the updated data group to the second non-volatile storage medium;

informing the peer controller that the updated data group is written to the second non-volatile storage medium;

discarding the updated data group from the peer controller upon being informed that the updated data group was successfully written to the second non-volatile storage medium;

detecting a system failure;

reading the updated data group from the second non-volatile storage medium and the comparison file into the controller memory;

recovering the previous data group from the updated data group and the comparison file using an XOR command; and

writing the recovered previous data group to the second non-volatile storage medium. 14.

9. The method of claim 8, the operations further comprising:

reading the updated data group from the medium and the comparison file into the controller memory;

recovering the previous data group from the updated data group and the comparison file; and

writing the recovered previous data group to the second non-volatile storage medium.

10. The method of claim 9, the operations further comprising:

determining to roll forward the updated data group;

recovering the updated data group from the previous data group and the comparison file using an XOR command; and

writing the recovered updated data group to the second non-volatile storage medium.