Using merkle trees in any point in time replication
One example method includes performing various operations while a stream of IOs is being applied to a source dataset and replicated to a target dataset, and the operations include receiving a replicated IO that was previously applied to a source dataset Merkle tree, applying the replicated IO to a target dataset Merkle tree by entering an updated hash value in the target dataset Merkle tree, and applying a timestamp to the target dataset Merkle tree.
1. A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors for:
performing, while a stream of Input/Output operations (IOs) is being applied to a source dataset at the same time as the source dataset is being replicated to a target dataset, operations comprising:
receiving a replicated IO that was previously applied to a source dataset tree;
applying the replicated IO to a target dataset tree by entering an updated hash value in the target dataset tree, wherein applying a timestamp to the target dataset tree comprises associating the timestamp with a root hash of the target dataset tree; and
applying a timestamp to the target dataset tree, wherein the timestamp is a single value; and
checking for consistency between the source dataset and the target dataset as of a specific Point-in Time (PiT).
2. The non-transitory storage medium as recited in claim 1 , wherein the source dataset tree and the target dataset tree each comprise a respective Merkle tree.
3. The non-transitory storage medium as recited in claim 1 , wherein each hash in the source dataset tree, and each hash in the target dataset tree, has a respective associated timestamp.
4. The non-transitory storage medium as recited in claim 1 , wherein the stream of IOs comprises one or more of a write operation and a delete operation.
5. The non-transitory storage medium as recited in claim 1 , wherein the replicated IO is a replication of an IO in the stream of IOs.
6. The non-transitory storage medium as recited in claim 1 , wherein the operations further comprise applying the stream of IOs to a source volume and to a target volume.
7. The non-transitory storage medium as recited in claim 1 , wherein the operations further comprise receiving, from a source, a value of a root hash of the source dataset tree.
8. The non-transitory storage medium as recited in claim 1 , wherein the source dataset tree comprises hashes of data blocks stored in a source volume, and the target dataset tree comprises hashes of data blocks stored in a target volume.
9. The non-transitory storage medium as recited in claim 1 , wherein the timestamp of the target dataset tree is the same as a timestamp of the most recent update to the target dataset tree.
10. A method, comprising:
performing, while a stream of Input/Output operations (IOs) is being applied to a source dataset at the same time as the source dataset is being replicated to a target dataset, operations comprising:
receiving a replicated IO that was previously applied to a source dataset tree;
applying the replicated IO to a target dataset tree by entering an updated hash value in the target dataset tree; and
applying a timestamp to the target dataset tree, wherein the timestamp is a single value, wherein applying a timestamp to the target dataset tree comprises associating the timestamp with a root hash of the target dataset tree; and
checking for consistency between the source dataset and the target dataset as of a specific Point-in Time (PiT).
11. The method as recited in claim 10 , wherein the source dataset tree and the target dataset tree each comprise a respective Merkle tree.
12. The method as recited in claim 10 , wherein each hash in the source dataset tree, and each hash in the target dataset tree, has a respective associated timestamp.
13. The method as recited in claim 10 , wherein the stream of IOs comprises one or more of a write operation and a delete operation.
14. The method as recited in claim 10 , wherein the replicated IO is a replication of an IO in the stream of IOs.
15. The method as recited in claim 10 , wherein the operations further comprise applying the stream of IOs to a source volume and to a target volume.
16. The method as recited in claim 10 , wherein the operations further comprise receiving, from a source, a value of a root hash of the source dataset tree.
17. The method as recited in claim 10 , wherein the source dataset tree comprises hashes of data blocks stored in a source volume, and the target dataset tree comprises hashes of data blocks stored in a target volume.
18. The method as recited in claim 10 , wherein the timestamp of the target dataset tree is the same as a timestamp of the most recent update to the target dataset tree.