Method to perform parallel data migration in a clustered storage environment

1. A method of migrating data from a source logical unit to a target logical unit, the source and target logical units corresponding to different areas of storage on one or more storage systems, the method comprising the steps of:

providing a plurality of nodes operational with the one or more storage systems; the plurality of nodes comprises a clustered system, the source logical unit and target logical unit are logical entities utilizing the clustered system to access the corresponding areas of storage;

providing by the nodes a LUN-device for access by a client, the LUN-device mapping to the source logical unit;

grouping data that are to be copied from the source logical unit to the target logical unit into chunks of data;

providing a bit-mask having one bit for each chunk of data;

dividing the bit-mask into splices of multiple bits; and

performing by each of the two or more of the plurality of nodes the following steps repeatedly until the source logical unit has been fully migrated to the target logical unit:

attempting to lock one of the splices;

if the splice is successfully locked, then:

copying the chunks of data corresponding to each bit in the splice to the target logical unit, wherein the copying of the chunks operates over the clustered system;

setting the bits in the bit-mask corresponding to the copied chunks; and

unlocking the splice when all the chunks corresponding to the bits in the splice have been copied;

migrating, in parallel by the two or more of the plurality of nodes, the data of the source logical unit to the target logical unit;

canceling the migrating at any time by coordinating the cancel request amongst all nodes in the cluster system, preserving the source logical unit and its corresponding LUN-device, and deleting the target logical unit and bit-mask;

storing multiple copies of the bit-mask in the clustered system, each copy of the bit-mask having an associated checksum; and

if any node of the plurality of nodes fails, checking by a surviving node the checksums to identify a valid copy of the bit-mask.

2. The method of claim 1 further comprising the step of: performing synchronous, sequential updates to those copies to ensure only one copy is being updated at any given time.

3. The method of claim 2 wherein the step of migrating completes successfully regardless of whether any of the plurality of nodes fails.

4. The method of claim 3 further comprising the steps of:

performing an I/O access by a node from a client to the LUN-device during the step of migrating, the step of performing comprising the steps of:

if the I/O access is a read access, returning data from the source logical unit to the client;

if the I/O access is a write access including data to be written:

ascertaining the chunk to which data is to be written:

locking the splice containing the chunk;

if all the bits in the splice are set, then writing the data to be written to the source logical unit and the target logical unit;

if less than all of the bits in the splice are set, then reading the chunks of data in the splice from the source logical unit, combining the data to be written with the chunks of data read, and writing the combined data to the target logical unit and source logical unit;

setting the bits in the bit mask corresponding to the copied chunks;

unlocking the splice.

5. Apparatus for migrating data from a source logical unit to a target logical unit, the source and target logical units corresponding to different areas of storage on one or more storage systems, the apparatus comprising:

a plurality of nodes operational with the one or more storage systems, the nodes comprising a clustered system, the source logical unit and target logical unit being logical entities that utilize the clustered system to access the corresponding areas of storage, the nodes providing a LUN-device for access by a client, the LUN-device mapping to the source logical unit, each node including logic for grouping data that are to be copied from the source logical unit to the target logical unit into chunks of data;

a bit-mask having one bit for each chunk;

logic for dividing the bit-mask into splices of multiple bits;

logic for attempting to lock one of the splices; and

logic for performing the following if the splice is successfully locked:

copying the chunks of data corresponding to each bit in the splice to the target logical unit, wherein the logic for copying operates over the clustered system;

setting the bits in the bit-mask corresponding to the copied chunks; and

unlocking the splice when all the chunks corresponding to the bits in the splice have been copied;

a mechanism on each node for causing two or more of the plurality of nodes to migrate in parallel the data of the source logical unit to the target logical unit, the mechanism including logic for canceling the migrating at any time, the logic for canceling including logic for coordinating the cancel request amongst all nodes in the cluster system, logic for preserving the source logical unit and its corresponding LUN-device, and logic for deleting the target logical unit and bit-mask;

logic for providing multiple copies of the bit-mask in the clustered system, each copy having an associated checksum; and

logic operational if any node of the plurality of nodes fails for checking by a surviving node the checksums to identify a valid copy of the bit-mask.

6. The apparatus of claim 5 further comprising logic for performing synchronous, sequential updates to those copies to ensure only one copy is being updated at any given time.

7. The apparatus of claim 6 wherein the source logical unit is successfully migrated to the target logical unit regardless of whether any of the plurality of nodes fails.

8. The apparatus of claim 7 further comprising:

logic performing an I/O access by a node from a client to the LUN-device during the step of migrating, the logic for performing comprising:

logic for returning data from the source logical unit to the client if the I/O access is a read access;

logic for performing the following if the I/O access is a write access including data to be written:

ascertaining the chunk to which data is to be written;

locking the splice containing the chunk;

if all the bits in the splice are set, then writing the data to be written to the source logical unit and the target logical unit;

if less than all of the bits in the splice are set, then reading the chunks of data in the splice from the source logical unit, combining the data to be written with the chunks of data read, and writing the combined data to the target logical unit and source logical unit;

setting the bits in the bit mask corresponding to the copied chunks;

unlocking the splice.

9. A program product comprising a computer readable medium having embodied therein a computer program for storing data, the computer program for operation on each of a plurality of nodes to migrate data from a source logical unit to a target logical unit, the source and target logical units corresponding to different areas of storage on one or more storage systems, the nodes comprising a clustered system, the source logical unit and target logical unit are logical entities utilizing the clustered system to access the corresponding areas of storage, the computer program comprising:

logic for providing a LUN-device for access by a client, the LUN-device mapping to the source logical unit;

logic for grouping data to be copied from the source logical unit to the target logical unit into chunks;

logic for providing a bit-mask having one bit for each chunk;

logic for dividing the bit-mask into splices of multiple bits; and

logic for attempting to lock one of the splices;

logic for performing includes the following if the splice is successfully locked:

logic for copying the chunks of data corresponding to each bit in the splice to the target logical unit, wherein the logic for copying the chunks operates over the clustered system;

logic for setting the bits in the bit-mask corresponding to the copied chunks; and

logic for unlocking the splice when all the chunks corresponding to the bits in the splice have been copied;

migration logic for migrating the data of the source logical unit to the target logical unit, the migration logic including logic for canceling the migrating at any time, the logic for canceling comprising logic for coordinating the cancel request amongst all nodes in the cluster, logic for preserving the source logical unit and its corresponding LUN-device, and logic for deleting the target logical unit and bit-mask, the migration logic operating in parallel with the migration logic on the other nodes such that each node participates in parallel in the migration of the data of the source logical unit;

logic for providing multiple copies of the bit-mask in the clustered system, each copy having associated with it a checksum; and

logic operational if any node of the plurality of nodes fails for checking by a surviving node the checksums to identify a valid copy of the bit mask.

10. The program product of claim 9 further comprising the logic for performing synchronous, sequential updates to those copies to ensure only one copy is being updated at any given time.

11. The program product of claim 10 wherein the source logical unit is successfully migrated to the target logical unit if any of the plurality of nodes fails.

12. The program product of claim 11 further comprising:

logic for performing an I/O access by a node from a client to the LUN-device during the migrating, the logic for performing comprising:

logic for returning data from the source logical unit to the client if the I/O access is a read access;

logic for performing the following if the I/O access is a write access including data to be written:

ascertaining the chunk to which data is to be written;

locking the splice containing the chunk;

if all the bits in the splice are set, then writing the data to be written to the source logical unit and the target logical unit;

if less than all of the bits in the splice are set, then reading the chunks of data in the splice from the source logical unit, combining the data to be written with the chunks of data read, and writing the combined data to the target logical unit and source logical unit;

setting the bits in the bit mask corresponding to the copied chunks;

unlocking the splice.