# US 2021/0241180 A1 · Automatically Allocating Device Resources 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 2021/0241180 A1 - Document ID: US20210241180A1 - Status: Published application - Kind: A1 - Application number: 16/778,508 - 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: Rahul Vishwakarma; Hemant Gaikwad; Gopal Singh - Web page: https://patentgrep.com/patent/US20210241180A1 ## CPC - G06F 11/2094: {Redundant storage or storage space (G06F11/2056 takes precedence)} - G06F 3/0617: {in relation to availability} - G06F 3/0619: {in relation to data integrity, e.g. data losses, bit errors} - G06F 3/0629: {Configuration or reconfiguration of storage systems} - G06F 3/0689: {Disk arrays, e.g. RAID, JBOD} - G06F 11/008: {Reliability or availability analysis} - G06F 11/2028: {eliminating a faulty processor or activating a spare} - G06F 18/2148: {characterised by the process organisation or structure, e.g. boosting cascade} - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 20/20: Ensemble learning - G06F 3/0631: {by allocating resources to storage systems} - G06F 11/0727: {in a storage system, e.g. in a DASD or network based storage system (drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units 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 11/0793: {Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40)} - 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 allocating device resources using machine learning techniques are provided herein. An example computer-implemented method includes determining values for multiple parameters from data obtained from one or more devices; generating at least one device component failure prediction by applying one or more machine learning techniques to at least a portion of the determined values; computing one or more qualifying values attributable to the at least one generated device component failure prediction by providing the at least one generated device component failure prediction to a conformal prediction framework; and automatically allocating one or more resources to at least a portion of the one or more devices based at least in part on the at least one generated device component failure prediction and the one or more computed qualifying values. ## Claims ### Claim 1 A computer-implemented method comprising: determining values for multiple parameters from data obtained from one or more devices; generating at least one device component failure prediction by applying one or more machine learning techniques to at least a portion of the determined values; computing one or more qualifying values attributable to the at least one generated device component failure prediction by providing the at least one generated device component failure prediction to a conformal prediction framework; and automatically allocating one or more resources to at least a portion of the one or more devices based at least in part on the at least one generated device component failure prediction and the one or more computed qualifying values; 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, further comprising: ranking each device component identified in the at least one generated device component failure prediction based at least in part on the one or more computed qualifying values. ### Claim 3 (depends on 2) The computer-implemented method of claim 2, further comprising: ranking a set of resources based at least in part on a level of interaction between each resource in the set of resources and each device component identified in the at least one generated device component failure prediction. ### Claim 4 (depends on 3) The computer-implemented method of claim 3, wherein automatically allocating the one or more resources comprises automatically allocating one or more resources from the set of resources based at least in part on the ranking of each device component identified in the at least one generated device component failure prediction and the ranking of the set of resources. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the one or more machine learning techniques comprises at least one stochastic gradient descent classification algorithm. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein the one or more machine learning techniques comprises at least one stochastic gradient descent classification algorithm integrated with at least one isotonic regression algorithm. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the conformal prediction framework comprises an inductive conformal prediction framework. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, further comprising: identifying the multiple parameters from a set of parameters by applying multiple machine learning techniques to a dataset related to device component failure. ### Claim 9 (depends on 8) The computer-implemented method of claim 8, wherein the multiple machine learning techniques comprise at least two of: one or more gradient boosting algorithms, one or more extra trees algorithms, one or more random forest algorithms, one or more decision tree algorithms, and one or more adaptive boosting algorithms. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, wherein the one or more devices comprise multiple storage devices, wherein the one or more resources comprises one or more spare disks, and wherein the computer-implemented method comprises: generating predictions pertaining to which of the multiple storage devices are going to fail by applying the one or more machine learning techniques to at least a portion of the determined values; computing confidence values attributable to the storage devices predicted to fail by providing the predictions to the conformal prediction framework; ranking the storage devices predicted to fail based at least in part on the computed confidence values; determining one or more redundant array of independent disk groups to which the disks predicted to fail belong; ranking the one or more redundant array of independent disk groups based at least in part on the disks predicted to fail contained therein; and automatically allocating one or more spare disks from a spare storage pool to at least one hot spare associated with the storage devices predicted to fail based at least in part on the ranking of the storage devices predicted to fail and the ranking of the one or more redundant array of independent disk groups. ### Claim 11 (depends on 1) The computer-implemented method of claim 1, wherein the one or more qualifying values comprise at least one confidence value attributed to the at least one generated device component failure prediction. ### Claim 12 (depends on 1) The computer-implemented method of claim 1, wherein the one or more qualifying values comprise at least one credibility value indicating quality of the data obtained from the one or more devices used in generating the at least one generated device component failure prediction. ### Claim 13 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 determine values for multiple parameters from data obtained from one or more devices; to generate at least one device component failure prediction by applying one or more machine learning techniques to at least a portion of the determined values; to compute one or more qualifying values attributable to the at least one generated device component failure prediction by providing the at least one generated device component failure prediction to a conformal prediction framework; and to automatically allocate one or more resources to at least a portion of the one or more devices based at least in part on the at least one generated device component failure prediction and the one or more computed qualifying values. ### Claim 14 (depends on 13) The non-transitory processor-readable storage medium of claim 13, wherein the program code when executed by the at least one processing device further causes the at least one processing device: to rank each device component identified in the at least one generated device component failure prediction based at least in part on the one or more computed qualifying values. ### Claim 15 (depends on 14) The non-transitory processor-readable storage medium of claim 14, wherein the program code when executed by the at least one processing device further causes the at least one processing device: to rank a set of resources based at least in part on a level of interaction between each resource in the set of resources and each device component identified in the at least one generated device component failure prediction. ### Claim 16 (depends on 15) The non-transitory processor-readable storage medium of claim 15, wherein automatically allocating the one or more resources comprises automatically allocating one or more resources from the set of resources based at least in part on the ranking of each device component identified in the at least one generated device component failure prediction and the ranking of the set of resources. ### 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 determine values for multiple parameters from data obtained from one or more devices; to generate at least one device component failure prediction by applying one or more machine learning techniques to at least a portion of the determined values; to compute one or more qualifying values attributable to the at least one generated device component failure prediction by providing the at least one generated device component failure prediction to a conformal prediction framework; and to automatically allocate one or more resources to at least a portion of the one or more devices based at least in part on the at least one generated device component failure prediction and the one or more computed qualifying values. ### Claim 18 (depends on 17) The apparatus of claim 17, the at least one processing device being further configured: to rank each device component identified in the at least one generated device component failure prediction based at least in part on the one or more computed qualifying values. ### Claim 19 (depends on 18) The apparatus of claim 18, the at least one processing device being further configured: to rank a set of resources based at least in part on a level of interaction between each resource in the set of resources and each device component identified in the at least one generated device component failure prediction. ### Claim 20 (depends on 19) The apparatus of claim 19, wherein automatically allocating the one or more resources comprises automatically allocating one or more resources from the set of resources based at least in part on the ranking of each device component identified in the at least one generated device component failure prediction and the ranking of the set of resources.