# US 2021/0240368 A1 · Automatically Determining Sizing Configurations for Storage Components Using Machine Learning Techniques > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0240368 A1 - Document ID: US20210240368A1 - Status: Published application - Kind: A1 - Application number: 16/778,553 - Filed: 2020-01-31 - Published: 2021-08-05 - Earliest claimed: 2020-01-31 - Assignees: EMC IP Holding Company LLC - Applicants: EMC IP Holding Company LLC - Inventors: Bina K. Thakkar; Deepak Gowda; Wenjin Liu - Web page: https://patentgrep.com/patent/US20210240368A1 ## CPC - G06F 3/0605: {by facilitating the interaction with a user or administrator} - G06F 3/0611: {in relation to response time} - G06F 3/0631: {by allocating resources to storage systems} - G06F 3/0641: {De-duplication techniques} - G06F 3/067: {Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]} - G06F 3/0689: {Disk arrays, e.g. RAID, JBOD} - G06F 18/24137: {Distances to cluster centroïds} - G06F 18/24147: {Distances to closest patterns, e.g. nearest neighbour classification} - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/0499: Feedforward networks - G06N 3/08: Learning methods - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Methods, apparatus, and processor-readable storage media for automatically determining sizing configurations for storage components using machine learning techniques are provided herein. An example computer-implemented method includes obtaining multiple items of input related to at least one storage component; determining a set of storage component sizing configurations by processing at least a portion of the multiple items of input using a first set of one or more machine learning techniques comprising at least one deep learning technique; identifying a subset of the storage component sizing configurations by processing at least a portion of the determined set of storage component sizing configurations using a second set of one or more machine learning techniques; and performing one or more automated actions based at least in part on the identified subset of storage component sizing configurations. ## Claims ### Claim 1 A computer-implemented method comprising: obtaining multiple items of input related to at least one storage component; determining a set of storage component sizing configurations by processing at least a portion of the multiple items of input using a first set of one or more machine learning techniques comprising at least one deep learning technique; identifying a subset of the storage component sizing configurations by processing at least a portion of the determined set of storage component sizing configurations using a second set of one or more machine learning techniques; and performing one or more automated actions based at least in part on the identified subset of storage component sizing configurations; wherein the method is performed by at least one processing device comprising a processor coupled to a memory. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the at least one deep learning technique comprises at least one neural network. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the at least one deep learning technique comprises at least one multilayer perceptron. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the one or more machine learning techniques in the second set comprises at least one machine learning-based k-nearest neighbors algorithm. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising: generating a dynamic workload performance library by applying at least one machine learning-based unsupervised clustering model to data pertaining to multiple storage components. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein the at least one machine learning-based unsupervised clustering model comprises at least one machine learning-based unsupervised clustering with Gaussian mixture model. ### Claim 7 (depends on 5) The computer-implemented method of claim 5, wherein the data pertaining to multiple storage components comprise configuration data and feature engineering data. ### Claim 8 (depends on 5) The computer-implemented method of claim 5, further comprising: updating the at least one deep learning technique using the generated dynamic workload performance library. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein the multiple items of inputs comprise information pertaining to two or more of input/output operations per second (IOPS), storage component capacity, at least one storage component workload, data reduction, deduplication, block size, drive type, and at least one redundant array of independent disks (RAID) group. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, wherein obtaining the multiple items of input comprises querying a storage systems database for data and applying at least one statistical analysis technique to the queried data. ### Claim 11 (depends on 1) The computer-implemented method of claim 1, wherein performing the one or more automated actions comprises outputting the identified subset of storage component sizing configurations to a user for selection of at least one of the storage component sizing configurations from the subset. ### Claim 12 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 obtain multiple items of input related to at least one storage component; to determine a set of storage component sizing configurations by processing at least a portion of the multiple items of input using a first set of one or more machine learning techniques comprising at least one deep learning technique; to identify a subset of the storage component sizing configurations by processing at least a portion of the determined set of storage component sizing configurations using a second set of one or more machine learning techniques; and to perform one or more automated actions based at least in part on the identified subset of storage component sizing configurations. ### Claim 13 (depends on 12) The non-transitory processor-readable storage medium of claim 12, wherein the at least one deep learning technique comprises at least one neural network. ### Claim 14 (depends on 12) The non-transitory processor-readable storage medium of claim 12, wherein the at least one deep learning technique comprises at least one multilayer perceptron. ### Claim 15 (depends on 12) The non-transitory processor-readable storage medium of claim 12, wherein the one or more machine learning techniques in the second set comprises at least one machine learning-based k-nearest neighbors algorithm. ### Claim 16 (depends on 12) The non-transitory processor-readable storage medium of claim 12, wherein the program code when executed by the at least one processing device further causes the at least one processing device: to generate a dynamic workload performance library by applying at least one machine learning-based unsupervised clustering model to data pertaining to multiple storage components; and to update the one or more deep learning techniques using the generated dynamic workload performance library. ### Claim 17 An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured: to obtain multiple items of input related to at least one storage component; to determine a set of storage component sizing configurations by processing at least a portion of the multiple items of input using a first set of one or more machine learning techniques comprising at least one deep learning technique; to identify a subset of the storage component sizing configurations by processing at least a portion of the determined set of storage component sizing configurations using a second set of one or more machine learning techniques; and to perform one or more automated actions based at least in part on the identified subset of storage component sizing configurations. ### Claim 18 (depends on 17) The apparatus of claim 17, wherein the at least one deep learning technique comprises at least one neural network. ### Claim 19 (depends on 17) The apparatus of claim 17, wherein the at least one deep learning technique comprises at least one multilayer perceptron. ### Claim 20 (depends on 17) The apparatus of claim 17, wherein the one or more machine learning techniques in the second set comprises at least one machine learning-based k-nearest neighbors algorithm.