Tagging write requests to avoid data-log bypass and promote inline deduplication during copies
View Patent ↗A storage technique includes selectively tagging a set of write requests by applying tags to write requests that are pursuant to copy operations and not applying tags to at least some write requests that are not pursuant to copy operations. For each of the set of write requests, the technique further includes performing a decision operation configured to (i) direct the write request down a first processing path that supports inline deduplication in response to the write request satisfying a set of criteria and (ii) direct the write request down a second processing path that does not support inline deduplication in response to the write request failing to satisfy the set of criteria. The set of criteria include whether the write request has a tag applied, such that the decision operation preferentially directs write requests down the first processing path in response to the write requests being tagged.
1. A method of performing writes in a data storage system, the method comprising:
selectively tagging a set of write requests by applying highly deduplicable (HD) tags to write requests that are pursuant to copy operations and not applying HD tags to at least some write requests that are not pursuant to copy operations;
for each of the set of write requests, performing a decision operation configured to (i) direct the write request down a first processing path that supports inline deduplication in response to the write request satisfying a set of criteria and (ii) direct the write request down a second processing path that does not support inline deduplication in response to the write request failing to satisfy the set of criteria, the set of criteria including whether the write request has an HD tag, such that the decision operation preferentially directs write requests down the first processing path in response to the write requests having HD tags; and
directing each of the set of write requests down one of the first processing path and the second processing path based on whether the respective write request satisfies the set of criteria.
2. The method of claim 1 , wherein directing each of the set of write requests down one of the first processing path and the second processing path includes directing a first write request down the first processing path by storing a first set of data specified by the first write request in a data log, the data log configured to flush the first set of data to a file system.
3. The method of claim 2 , wherein directing each of the set of write requests down one of the first processing path and the second processing path includes directing a second write request down the second processing path by bypassing the data log and storing a second set of data specified by the second write request in the file system, without storing the second set of data in the data log.
4. The method of claim 3 , wherein the set of criteria further includes a fullness metric that describes a level of fullness of the data log, and wherein the method further comprises the decision operation directing write requests down the second processing path based at least in part on the fullness metric.
5. The method of claim 4 , further comprising:
the decision operation directing an HD-tagged write request down the first processing path when the data log has a particular value of the fullness metric; and
the decision operation directing a non-HD-tagged write request down the second processing path when the data log has the same particular value of the fullness metric.
6. The method of claim 3 , wherein the data log is backed by a cache that includes multiple tracks, each track configured to prevent eviction of data for a respective minimum amount of time, the minimum amount of time for a first track being greater than the minimum amount of time for a second track, and wherein the method further comprises, when storing data specified by write requests in the data log, storing data specified by HD-tagged write requests in the first track of the cache.
7. The method of claim 3 , further comprising, when flushing the first set of data from the data log to the file system, performing an inline deduplication on the first set of data to avoid creating a duplicate of the first set of data in the file system.
8. A computerized apparatus, comprising control circuitry that includes a set of processing units coupled to memory, the control circuitry constructed and arranged to:
selectively tag a set of write requests by applying highly deduplicable (HD) tags to write requests that are pursuant to copy operations and not applying HD tags to at least some write requests that are not pursuant to copy operations;
for each of the set of write requests, perform a decision operation configured to (i) direct the write request down a first processing path that supports inline deduplication in response to the write request satisfying a set of criteria and (ii) direct the write request down a second processing path that does not support inline deduplication in response to the write request failing to satisfy the set of criteria, the set of criteria including whether the write request has an HD tag, such that the decision operation preferentially directs write requests down the first processing path in response to the write requests having HD tags; and
direct each of the set of write requests down one of the first processing path and the second processing path based on whether the respective write request satisfies the set of criteria.
9. The computerized apparatus of claim 8 , wherein the control circuitry constructed and arranged to direct each of the set of write requests down one of the first processing path and the second processing path is further constructed and arranged to direct a first write request down the first processing path by storing a first set of data specified by the first write request in a data log, the data log configured to flush the first set of data to a file system.
10. The computerized apparatus of claim 9 , wherein the control circuitry constructed and arranged to direct each of the set of write requests down one of the first processing path and the second processing path is further constructed and arranged to direct a second write request down the second processing path by bypassing the data log and storing a second set of data specified by the second write request in the file system, without storing the second set of data in the data log.
11. The computerized apparatus of claim 10 , wherein the set of criteria further includes a fullness metric that describes a level of fullness of the data log, and wherein the decision operation is further configured to direct write requests down the second processing path based at least in part on the fullness metric.
12. The computerized apparatus of claim 11 ,
wherein the decision operation is further configured to direct an HD-tagged write request down the first processing path when the data log has a particular value of the fullness metric; and
wherein the decision operation is further configured to direct a non-HD-tagged write request down the second processing path when the data log has the same particular value of the fullness metric.
13. The computerized apparatus of claim 10 , wherein the data log is backed by a cache that includes multiple tracks, each track configured to prevent eviction of data for a respective minimum amount of time, the minimum amount of time for a first track being greater than the minimum amount of time for a second track, and wherein, when constructed and arranged to store data specified by write requests in the data log, the control circuitry is further constructed and arranged to store data specified by HD-tagged write requests in the first track of the cache.
14. A computer program product including a set of non-transitory, computer-readable media having instructions which, when executed by control circuitry of a computerized apparatus, cause the control circuitry to perform a method for copying data, the method comprising:
selectively tagging a set of write requests by applying highly deduplicable (HD) tags to write requests that are pursuant to copy operations and not applying HD tags to at least some write requests that are not pursuant to copy operations;
for each of the set of write requests, performing a decision operation configured to (i) direct the write request down a first processing path that supports inline deduplication in response to the write request satisfying a set of criteria and (ii) direct the write request down a second processing path that does not support inline deduplication in response to the write request failing to satisfy the set of criteria, the set of criteria including whether the write request has an HD tag, such that the decision operation preferentially directs write requests down the first processing path in response to the write requests having HD tags; and
directing each of the set of write requests down one of the first processing path and the second processing path based on whether the respective write request satisfies the set of criteria.
15. The computer program product of claim 14 , wherein directing each of the set of write requests down one of the first processing path and the second processing path includes directing a first write request down the first processing path by storing a first set of data specified by the first write request in a data log, the data log configured to flush the first set of data to a file system.
16. The computer program product of claim 15 , wherein directing each of the set of write requests down one of the first processing path and the second processing path includes directing a second write request down the second processing path by bypassing the data log and storing a second set of data specified by the second write request in the file system, without storing the second set of data in the data log.
17. The computer program product of claim 16 , wherein the set of criteria further includes a fullness metric that describes a level of fullness of the data log, and wherein the method further comprises the decision operation directing write requests down the second processing path based at least in part on the fullness metric.
18. The computer program product of claim 17 , wherein the method further comprises:
the decision operation directing an HD-tagged write request down the first processing path when the data log has a particular value of the fullness metric; and
the decision operation directing a non-HD-tagged write request down the second processing path when the data log has the same particular value of the fullness metric.
19. The computer program product of claim 16 , wherein the data log is backed by a cache that includes multiple tracks, each track configured to prevent eviction of data for a respective minimum amount of time, the minimum amount of time for a first track being greater than the minimum amount of time for a second track, and wherein the method further comprises, when storing data specified by write requests in the data log, storing data specified by HD-tagged write requests in the first track of the cache.
20. The computer program product of claim 16 , wherein the method further comprises, when flushing the first set of data from the data log to the file system, performing an inline deduplication on the first set of data to avoid creating a duplicate of the first set of data in the file system.