Method and system for garbage collection in a storage system based on longevity of stored data
View Patent ↗A method for managing data. The method includes receiving a first request to write data to persistent storage and in response to the first request, writing the data to a short-lived block in the persistent storage, where the data is short-lived data or data of unknown longevity. The method further includes performing a modified garbage collection operation that includes: selecting a first frag page in a first block, determining that the first frag page is live, and migrating, based on the determination that the first frag page is live, the first frag page to a long-lived block in the persistent storage, where the long-lived block is distinct from the short-lived block and wherein the long-lived block does not include any short-lived data.
1. A non-transitory computer readable medium (CRM) comprising instructions, which when executed by a processor perform a method, the method comprising:
selecting a frag page of a block stored on a persistent storage;
generating a derived physical address using a TOC entry associated with the frag page;
selecting a stored physical address based on an object ID and an offset ID, wherein the object ID and the offset ID are located in the TOC entry;
making a determination, that the derived physical address matches the stored physical address; and
in response to the determination, migrating the frag page to a second block stored on the persistent storage.
2. The non-transitory CRM of claim 1 , wherein the method further comprises:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address does not match the second stored physical address; and
in response to the second determination, deleting the second frag page without migrating the second frag page.
3. The non-transitory CRM of claim 2 , wherein the method further comprises:
while deleting the second frag page, deleting the first frag page.
4. The non-transitory CRM of claim 3 , wherein the first frag page and the second frag page are deleted as a block level deletion of the block.
5. The non-transitory CRM of claim 1 , wherein the method further comprises:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address matches the second stored physical address; and
in response to the second determination, migrating the second frag page to the second block stored on the persistent storage.
6. The non-transitory CRM of claim 5 , wherein the method further comprises:
making a third determination that a first candidate block has a lower wear level than a wear level of a second candidate block; and
selecting the first candidate block as the second block.
7. The non-transitory CRM of claim 5 , wherein the method further comprises:
making a third determination that a candidate block is empty; and
selecting the candidate block as the second block.
8. The non-transitory CRM of claim 1 , wherein the method further comprises:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address matches the second stored physical address; and
in response to the second determination, migrating the second frag page to a third block stored on the persistent storage.
9. A method for managing data, comprising:
selecting a frag page of a block stored on a persistent storage;
generating a derived physical address using a TOC entry associated with the frag page;
selecting a stored physical address based on an object ID and an offset ID, wherein the object ID and the offset ID are located in the TOC entry;
making a determination, that the derived physical address matches the stored physical address; and
in response to the determination, migrating the frag page to a second block stored on the persistent storage.
10. The method of claim 9 , further comprising:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address does not match the second stored physical address; and
in response to the second determination, deleting the second frag page without migrating the second frag page.
11. The method of claim 10 , wherein the method further comprises:
while deleting the second frag page, deleting the first frag page.
12. The method of claim 11 , wherein the first frag page and the second frag page are deleted as a block level deletion of the block.
13. The method of claim 9 , further comprising:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address matches the second stored physical address; and
in response to the second determination, migrating the second frag page to the second block stored on the persistent storage.
14. The method of claim 13 , further comprising:
making a third determination that a first candidate block has a lower wear level than a wear level of a second candidate block; and
selecting the first candidate block as the second block.
15. The method of claim 13 , further comprising:
making a third determination that a candidate block is empty; and
selecting the candidate block as the second block.
16. The method of claim 9 , further comprising:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address matches the second stored physical address; and
in response to the second determination, migrating the second frag page to a third block stored on the persistent storage.
17. A storage appliance, comprising:
persistent storage;
a non-transitory computer readable medium comprising instructions;
a processor, configured to execute the instructions, wherein the instructions when executed by the processor perform a method, the method comprising:
selecting a frag page of a block stored on the persistent storage;
generating a derived physical address using a TOC entry associated with the frag page;
selecting a stored physical address based on an object ID and an offset ID, wherein the object ID and the offset ID are located in the TOC entry;
making a determination, that the derived physical address matches the stored physical address; and
in response to the determination, migrating the frag page to a second block stored on the persistent storage.
18. The storage appliance of claim 17 , wherein the method further comprises:
selecting a second frag page of the block stored on the persistent storage;
generating a second derived physical address using a second TOC entry associated with the second frag page;
selecting a second stored physical address based on a second object ID and a second offset ID, wherein the second object ID and the second offset ID are located in the second TOC entry;
making a second determination, that the second derived physical address does not match the second stored physical address; and
in response to the second determination, deleting the second frag page without migrating the second frag page.
19. The storage appliance of claim 18 , wherein the method further comprises:
while deleting the second frag page, deleting the first frag page.
20. The storage appliance of claim 19 , wherein the first frag page and the second frag page are deleted as a block level deletion of the block.