# US 11,442,650 B2 · Generating predicted usage of storage capacity > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,442,650 B2 - Document ID: US11442650B2 - Status: Granted - Kind: B2 - Application number: 16/825,065 - Filed: 2020-03-20 - Published: 2020-11-05 - Granted: 2022-09-13 - Earliest claimed: 2019-04-30 - Assignees: EMC IP Holding Company LLC - Applicants: EMC IP Holding Company LLC - Inventors: Jie Huang; Xudong Wang; Ming Wang; Xiaoyu Ren - Web page: https://patentgrep.com/patent/US11442650B2 ## CPC - G06F 3/0649: {Lifecycle management} - G06F 3/0604: {Improving or facilitating administration, e.g. storage management} - G06F 3/0613: {in relation to throughput} - G06F 3/0616: {in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]} - G06F 3/0653: {Monitoring storage devices or systems} - G06F 3/0673: {Single storage device} - G06F 11/3034: {where the computing system component is a storage system, e.g. DASD based or network based (digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097)} - G06F 11/3409: {for performance assessment} - G06F 11/3447: {Performance evaluation by modeling} - G06F 11/3466: {Performance evaluation by tracing or monitoring} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Storage management techniques involve: obtaining a historical usage of storage capacity for a storage device, and a historical feature characterizing the historical usage of storage capacity; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. Therefore, such techniques can accurately predict the usage of storage capacity for the storage device. ## Claims ### Claim 1 A method for storage management, comprising: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval, and a timestamp associated with the predetermined time interval, wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 2 (depends on 1) The method of claim 1, wherein generating the predicted usage of storage capacity comprises: normalizing the historical feature; and generating the predicted usage of storage capacity by inputting the normalized historical feature to the predictor. ### Claim 3 (depends on 1) The method of claim 1, wherein comparing the historical usage of storage capacity with the predicted usage of storage capacity comprises: normalizing the historical usage of storage capacity; and comparing the normalized historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 4 (depends on 1) The method of claim 1, wherein the predictor is a Long Short-Term Memory (LSTM) neural network. ### Claim 5 (depends on 1) The method of claim 1, wherein the historical feature further comprises: an average number of write operations per second for the storage device within the predetermined time interval. ### Claim 6 (depends on 1) The method of claim 1, further comprising: obtaining a further historical usage of storage capacity for the storage device, and a further historical feature characterizing the further historical usage of storage capacity; generating a further predicted usage of storage capacity for the storage device based on the further historical feature and the updated predictor; and determining performance of the predictor by comparing the further historical usage of storage capacity and the further predicted usage of storage capacity. ### Claim 7 A method for storage management, comprising: obtaining a feature characterizing a usage of storage capacity for a storage device; and predicting the usage of storage capacity for the storage device based on the feature and a predictor updated by a method of storage management which includes: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval and a timestamp associated with the predetermined time interval, and wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 8 (depends on 7) The method of claim 7, wherein the feature further comprises: an average number of write operations per second for the storage device within the predetermined time interval. ### Claim 9 A device for storage management, comprising: at least one processing unit; at least one memory coupled to the at least one processing unit and storing instructions executed by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the device to perform acts comprising: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval and a timestamp associated with the predetermined time interval, and wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 10 (depends on 9) The device of claim 9, wherein generating the predicted usage of storage capacity comprises: normalizing the historical feature; and generating the predicted usage of storage capacity by inputting the normalized historical feature to the predictor. ### Claim 11 (depends on 9) The device of claim 9, wherein comparing the historical usage of storage capacity with the predicted usage of storage capacity comprises: normalizing the historical usage of storage capacity; and comparing the normalized historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 12 (depends on 9) The device of claim 9, wherein the predictor is a Long Short-Term Memory (LSTM) neural network. ### Claim 13 (depends on 9) The device of claim 9, wherein the historical feature further comprises: an average number of write operations per second for the storage device within the predetermined time interval. ### Claim 14 (depends on 9) The device of claim 9, the acts further comprising: obtaining a further historical usage of storage capacity for the storage device, and a further historical feature characterizing the further historical usage of storage capacity; generating a further predicted usage of storage capacity for the storage device based on the further historical feature and the updated predictor; and determining performance of the predictor by comparing the further historical usage of storage capacity and the further predicted usage of storage capacity. ### Claim 15 A device for storage management, comprising: at least one processing unit; at least one memory coupled to the at least one processing unit and storing instructions executed by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the device to perform acts comprising: obtaining a feature characterizing a usage of storage capacity for a storage device; and predicting the usage of storage capacity for the storage device based on the feature and a predictor updated by a method of storage management which includes: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval and a timestamp associated with the predetermined time interval, and wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 16 (depends on 15) The device of claim 15, wherein the feature further comprises: an average number of write operations per second for the storage device within the predetermined time interval. ### Claim 17 A computer program product having a non-transitory computer readable medium which stores a set of instructions perform storage management; the set of instructions, when carried out by computerized circuitry, causing the computerized circuitry to perform a method of: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval and a timestamp associated with the predetermined time interval, and wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity. ### Claim 18 A computer program product having a non-transitory computer readable medium which stores a set of instructions to perform storage management; the set of instructions, when carried out by computerized circuitry, causing the computerized circuitry to perform a method of: obtaining a feature characterizing a usage of storage capacity for a storage device; and predicting the usage of storage capacity for the storage device based on the feature and a predictor updated by a method of storage management which includes: obtaining a historical usage of storage capacity for a storage device; obtaining, for the storage device, raw data associated with the historical usage of storage capacity; determining, as a historical feature characterizing the historical usage of storage capacity, a feature in the raw data, responsive to a relevance of the feature to the historical usage of storage capacity exceeding at least one predetermined threshold, wherein the historical feature comprises an average size of data written for the storage device within a predetermined time interval, a physical space size for the storage device used within the predetermined time interval and a timestamp associated with the predetermined time interval, and wherein the timestamp indicates a sampling time point associated with the predetermined time interval; generating a predicted usage of storage capacity for the storage device based on the historical feature and a predictor for predicting a usage of storage capacity; and updating the predictor by comparing the historical usage of storage capacity with the predicted usage of storage capacity.