Storage system configured for efficient generation of capacity release estimates for deletion of datasets
A storage system in one embodiment comprises a plurality of storage devices and an associated storage controller. The storage controller is configured to identify a dataset to be scanned to generate a capacity release estimate for prospective deletion of that dataset, to designate a content-based signature prefix to be utilized in the scan, and to scan logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix. The scanning further comprises, for each such identified page, determining a reference count of the page, and responsive to the reference count of the page having a particular value, determining a compressibility measure for the page. The storage controller generates the capacity release estimate for prospective deletion of the dataset based at least in part on the one or more page compressibility measures determined as part of the scan.
1. An apparatus comprising:
a storage system comprising a plurality of storage devices and an associated storage controller;
the storage controller being configured:
to identify a dataset to be scanned to generate a capacity release estimate for prospective deletion of that dataset;
to designate a content-based signature prefix to be utilized in the scan;
to scan logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix, wherein the scanning further comprises, for each such identified page:
determining a reference count of the page, the reference count of the page being distinct from but maintained for a corresponding content-based signature of the page, such that pages having the same content-based signature have the same reference count; and
responsive to the reference count of the page having a particular value, determining a compressibility measure for the page; and
to generate the capacity release estimate for prospective deletion of the dataset based at least in part on the one or more page compressibility measures;
wherein the reference count of the page specifies a total number of pages that have the same content-based signature and that map via that same content-based signature to the same portion of a physical storage space of one or more of the storage devices;
wherein the particular value of the reference count of the page that is utilized to control whether or not a compressibility measure is determined for that page comprises a reference count value of one; and
wherein the storage controller comprises at least one processing device comprising a processor coupled to a memory.
2. The apparatus of claim 1 wherein the dataset comprises a particular point-in-time snapshot of a set of one or more logical storage volumes and wherein each of the logical storage volumes comprises at least a portion of the physical storage space of the one or more of the storage devices.
3. The apparatus of claim 1 wherein scanning logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix comprises:
identifying an address-to-hash mapping table for the dataset; and
traversing the address-to-hash mapping table to identify one or more entries each including a hash value having the designated content-based signature prefix.
4. The apparatus of claim 3 wherein the hash values in respective entries of the address-to-hash mapping table comprise at least respective portions of corresponding content-based signatures comprising hash digests of respective ones of the pages.
5. The apparatus of claim 3 wherein the hash values in respective entries of the address-to-hash mapping table comprise respective hash handles corresponding to respective content-based signatures of respective ones of the pages.
6. The apparatus of claim 3 wherein a sampling ratio for the scan is a function of a length of the content-based signature prefix in bytes.
7. The apparatus of claim 1 wherein determining a compressibility measure for the page comprises reading a stored compressibility measure generated in conjunction with a previous compression of the page.
8. The apparatus of claim 1 wherein determining a compressibility measure for the page comprises:
retrieving the page;
compressing the page; and
generating the compressibility measure based at least in part on the compressing.
9. The apparatus of claim 1 wherein the designated content-based signature prefix comprises a specified number of initial content-based signatures bytes each having a designated value.
10. The apparatus of claim 9 wherein the content-based signature prefix comprises a single byte.
11. The apparatus of claim 1 wherein generating the capacity release estimate for the dataset based at least in part on the one or more page compressibility measures further comprises:
computing a partial capacity release estimate based at least in part on multiple page compressibility measures used to determine compressed page sizes for respective ones of the pages; and
scaling the partial capacity release estimate as a function of a sampling ratio to obtain the capacity release estimate for the dataset.
12. The apparatus of claim 1 wherein the storage controller is configured:
to generate one or more additional capacity release estimates for respective ones of one or more additional datasets; and
to select a particular one of the datasets for deletion based at least in part on their respective capacity release estimates.
13. A method comprising:
identifying in a storage system a dataset to be scanned to generate a capacity release estimate for prospective deletion of that dataset;
designating a content-based signature prefix to be utilized in the scan;
scanning logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix, wherein the scanning further comprises, for each such identified page:
determining a reference count of the page, the reference count of the page being distinct from but maintained for a corresponding content-based signature of the page, such that pages having the same content-based signature have the same reference count; and
responsive to the reference count of the page having a particular value, determining a compressibility measure for the page; and
generating the capacity release estimate for prospective deletion of the dataset based at least in part on the one or more page compressibility measures;
wherein the reference count of the page specifies a total number of pages that have the same content-based signature and that map via that same content-based signature to the same portion of a physical storage space of one or more of storage devices of the storage system;
wherein the particular value of the reference count of the page that is utilized to control whether or not a compressibility measure is determined for that page comprises a reference count value of one; and
wherein the method is implemented by at least one processing device comprising a processor coupled to a memory.
14. The method of claim 13 wherein scanning logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix comprises:
identifying an address-to-hash mapping table for the dataset; and
traversing the address-to-hash mapping table to identify one or more entries each including a hash value having the designated content-based signature prefix.
15. The method of claim 13 wherein the designated content-based signature prefix comprises a specified number of initial content-based signatures bytes each having a designated value.
16. The method of claim 13 further comprising:
generating one or more additional capacity release estimates for respective ones of one or more additional datasets; and
selecting a particular one of the datasets for deletion based at least in part on their respective capacity release estimates.
17. A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:
to identify in a storage system a dataset to be scanned to generate a capacity release estimate for prospective deletion of that dataset;
to designate a content-based signature prefix to be utilized in the scan;
to scan logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix, wherein the scanning further comprises, for each such identified page:
determining a reference count of the page, the reference count of the page being distinct from but maintained for a corresponding content-based signature of the page, such that pages having the same content-based signature have the same reference count; and
responsive to the reference count of the page having a particular value, determining a compressibility measure for the page; and
to generate the capacity release estimate for prospective deletion of the dataset based at least in part on the one or more page compressibility measures;
wherein the reference count of the page specifies a total number of pages that have the same content-based signature and that map via that same content-based signature to the same portion of a physical storage space of one or more storage devices of the storage system; and
wherein the particular value of the reference count of the page that is utilized to control whether or not a compressibility measure is determined for that page comprises a reference count value of one.
18. The computer program product of claim 17 wherein scanning logical address mapping information for the dataset to identify one or more pages of the dataset that have the designated content-based signature prefix comprises:
identifying an address-to-hash mapping table for the dataset; and
traversing the address-to-hash mapping table to identify one or more entries each including a hash value having the designated content-based signature prefix.
19. The computer program product of claim 17 wherein the designated content-based signature prefix comprises a specified number of initial content-based signatures bytes each having a designated value.
20. The computer program product of claim 17 wherein the program code when executed by the at least one processing device causes the at least one processing device:
to generate one or more additional capacity release estimates for respective ones of one or more additional datasets; and
to select a particular one of the datasets for deletion based at least in part on their respective capacity release estimates.