EFFICIENT METHOD AND APPARATUS FOR KEEPING TRACK OF IN FLIGHT DATA IN A DUAL NODE STORAGE CONTROLLER

• US 20100146348A1
• Filed: 12/08/2008
• Published: 06/10/2010
• Est. Priority Date: 12/08/2008
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Zlotnick, Aviad
• Granted Patent: US 8,176,363 B2
• Time in Patent Office: Days
• US Class Current: 714/54
• CPC Class Codes: G06F 11/2082 Data synchronisation
• CPC Class Codes: G06F 11/2087 with a common controller

Abstract

A storage unit adapted for use in a processing system includes a controller including at least two compute nodes, each of the compute nodes including a read/write cache and a persistent cache; and machine executable instructions stored within machine readable media, the instructions including instructions for tracking in-flight data in the persistent cache and composing a list of the in-flight data segments after a failure of the controller. A processing system and a method are also provided.

Claims

1. A storage unit adapted for use in a processing system, the storage unit comprising:

a controller comprising at least two compute nodes, each of the compute nodes comprising a read/write cache and a persistent cache; and

machine executable instructions stored within machine readable media, the instructions comprising instructions for tracking in-flight data in the persistent cache and composing a list of the in-flight data segments after a failure of the controller.

2. The storage unit of claim 1, wherein tracking comprises setting a bit to indicate the in-flight data.

3. The storage unit of claim 1, wherein tracking comprises maintaining a copy of the in-flight data in the persistent storage.

4. The storage unit of claim 1, wherein the instructions further comprise instructions for sending a message upon completion of a successful transfer.

5. The storage unit of claim 1, wherein the instructions further comprise instructions for reading the list and recovering the in-flight data segments after the failure.

6. The storage unit of claim 5, wherein the instructions further comprise instructions for copying list data from another storage unit.

7. The storage unit of claim 1, wherein the machine readable media comprises at least one of backend storage, non-volatile storage and local read-only-memory.

8. A processing system comprising:

a plurality of mass storage units, at least a first one of the mass storage units comprising a controller comprising at least two compute nodes, each of the compute nodes comprising a read/write cache and a persistent cache; and

machine executable instructions stored within machine readable media, the instructions comprising instructions for tracking in-flight data in the persistent cache and composing a list of the in-flight data segments after a failure of the controller.

9. The processing system of claim 8, wherein the instructions further comprise instructions for copying list data from another storage unit after the failure.

10. A method for storing data in a local mass storage unit, the method comprising:

storing data in a cache of first node of a dual-node controller of the local mass storage unit;

copying the data into non-volatile storage of a second node of the controller;

sending the data to a remote controller for storage in a remote mass storage unit;

waiting for the data to be stored in the local mass storage unit and the remote mass storage unit; and

discarding the data in the non-volatile storage upon local and remote writing of the data.

11. The method as in claim 10, further comprising:

after a failure of the dual-node controller and during a recovery of the local mass storage unit, getting a list of data segments in the non-volatile storage of the second node; and

copying contents of the data segments from the remote mass storage unit to the local mass storage unit.

12. The method as in claim 10, wherein waiting comprises testing for completion of a write to the remote mass storage unit.

13. The method as in claim 10, wherein waiting comprises testing for completion of a write to the local mass storage unit.