US 12,013,816 B2
Querying metadata in a storage system
Sining Wu, Chichester (GB)
Assigned to Seagate Technology LLC, Fremont, CA (US)
Filed by Seagate Technology LLC, Fremont, CA (US)
Filed on Jan. 28, 2022, as Appl. No. 17/587,281.
Prior Publication US 2023/0244637 A1, Aug. 3, 2023
Int. Cl. G06F 16/14 (2019.01); G06F 16/13 (2019.01); G06F 16/16 (2019.01)
CPC G06F 16/156 (2019.01) [G06F 16/137 (2019.01); G06F 16/164 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method of efficiently querying metadata in a storage system, the method comprising:
receiving a metadata query from a client node that asks one or more questions regarding a set of metadata in a metadata store;
executing the metadata query on at least one metadata sketch, wherein the at least one metadata sketch summarizes a metadata stream of metadata access during a period of time and thereby has a smaller memory footprint than the metadata stream during the period of time, and wherein different metadata sketches from different periods of time are periodically merged together so that the metadata query is conducted on a merged metadata sketch;
returning an answer from the metadata query executed on the merged metadata sketch, wherein the answer is an approximate response, which is within a predetermined bounded distance from an actual answer, to the one or more questions regarding the set of metadata in the metadata store, and wherein the predetermined bounded distance encompasses a portion of the merged metadata sketch that comprises a data size that is less than a data size of the merged metadata sketch; and
performing analysis of the storage system or data stored in the storage system with the approximate response.