Capacity determination for content-based storage
View Patent ↗A computer program product, system, and method for visiting each node of a snapshot tree within a content-based storage system having a plurality of volumes and/or snapshots; for each node, scanning an address-to-hash (A2H) table to calculate one or more resource usage metrics, wherein the A2H tables map logical I/O addresses to chunk hashes; and determining, based on the resource usage metrics, an amount of memory and/or disk capacity that would be freed by deleting one or more of the volumes and/or snapshots.
1. A method comprising:
visiting each node of a snapshot tree within a content-based storage system having a plurality of volumes and snapshots;
for each node, scanning an address-to-hash (A2H) table to calculate one or more resource usage metrics, wherein the A2H tables map logical I/O addresses to chunk hashes;
determining, based on the resource usage metrics, an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots, the determining performed on a per-node basis and including determining the amount of memory for each ancestor node of the node that is not also an ancestor node of another node, determining the amount of memory for each leaf node of the node, and summing the resource usage metrics for each of the ancestor nodes of the node and for each of the leaf nodes of the node; and
deleting the one or more of the volumes and snapshots from the content-based storage system determined based on the per-node resource usage metrics.
2. The method of claim 1 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes determining a count of entries in the A2H table.
3. The method of claim 2 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, an amount of memory based on the count of entries in the A2H table.
4. The method of claim 2 further comprising determining accessible space provided by one or more of the volumes and snapshots.
5. The method of claim 1 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes:
finding one or more unique chunks associated with the node; and
determining a compression ratio for each of the unique chunks associated with the node.
6. The method of claim 5 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, a minimum disk capacity that would be freed by deleting the node using a count of unique chunks and the compression ratios.
7. The method of claim 1 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes:
determining a count of chunks associated with the node;
determining a compression ratio for each of the chunks associated with the node; and
determining a reference count for each of the chunks associated with the node.
8. The method of claim 7 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, an estimated disk capacity that would be freed by deleting the node based on the count of chunks associated with the node, the compression ratios, and the reference counts.
9. A computer program product tangibly embodied in a non-transitory computer-readable medium, the computer-readable medium storing program instructions that are executable to:
visit each node of a snapshot tree within a content-based storage system having a plurality of volumes and snapshots;
for each node, scan an address-to-hash (A2H) table to calculate one or more resource usage metrics, wherein the A2H tables map logical I/O addresses to chunk hashes;
determine, based on the resource usage metrics, an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots, the determining performed on a per-node basis and including determining the amount of memory for each ancestor node of the node that is not also an ancestor node of another node, determining the amount of memory for each leaf node of the node, and summing the resource usage metrics for each of the ancestor nodes of the node and for each of the leaf nodes of the node; and
delete the one or more of the volumes and snapshots from the content-based storage system determined based on the per-node resource usage metrics.
10. The computer product of claim 9 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes determining a count of entries in the A2H table.
11. The computer product of claim 10 wherein determining an one or more of amount of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, an amount of memory based on the count of entries in the A2H table.
12. The computer product of claim 10 further comprising
determining accessible space provided by one or more of the volumes and snapshots.
13. A system comprising:
a processor;
a volatile memory; and
a non-volatile memory storing computer program code that when executed on the processor causes the processor to execute a process operable to:
visit each node of a snapshot tree within a content-based storage system having a plurality of volumes and snapshots;
for each node, scan an address-to-hash (A2H) table to calculate one or more resource usage metrics, wherein the A2H tables map logical I/O addresses to chunk hashes;
determine, based on the resource usage metrics, an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots, the determining performed on a per-node basis and including determining the amount of memory for each ancestor node of the node that is not also an ancestor node of another node, determining the amount of memory for each leaf node of the node, and summing the resource usage metrics for each of the ancestor nodes of the node and for each of the leaf nodes of the node; and
delete the one or more of the volumes and snapshots from the content-based storage system determined based on the per-node resource usage metrics.
14. The system of claim 13 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes determining a count of entries in the A2H table.
15. The system of claim 14 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, an amount of memory based on the count of entries in the A2H table.
16. The system of claim 14 further comprising determining accessible space provided by one or more of the volumes and snapshots.
17. The system of claim 13 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes:
finding one or more unique chunks associated with the node; and
determining a compression ratio for each of the unique chunks associated with the node.
18. The system of claim 17 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, a minimum disk capacity that would be freed by deleting the node using a count of unique chunks and the compression ratios.
19. The system of claim 13 wherein, for each node, scanning the A2H table to calculate one or more resource usage metrics includes:
determining a count of chunks associated with the node;
determining a compression ratio for each of the chunks associated with the node; and
determining a reference count for each of the chunks associated with the node.
20. The system of claim 19 wherein determining an amount of one or more of memory and disk capacity that would be freed by deleting one or more of the volumes and snapshots includes determining, for each node, an estimated disk capacity that would be freed by deleting the node based on the count of chunks associated with the node, the compression ratios, and the reference counts.