# US 12,008,418 B2 · Automated causal analysis of issues affecting workloads executing in an information technology infrastructure > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,008,418 B2 - Document ID: US12008418B2 - Status: Granted - Kind: B2 - Application number: 17/462,710 - Filed: 2021-08-31 - Published: 2023-03-02 - Granted: 2024-06-11 - Earliest claimed: 2021-08-31 - Assignees: Dell Products L.P. - Applicants: Dell Products L.P. - Inventors: Vindhya Gangaraju; Vinay Sawal; Jayanth Kumar Reddy Perneti - Web page: https://patentgrep.com/patent/US12008418B2 ## CPC - G06F 9/5083: {Techniques for rebalancing the load in a distributed system} - G06F 11/1446: Point-in-time backing up or restoration of persistent data - G06F 11/1448: {Management of the data involved in backup or backup restore} - G06F 2201/805: Real-time - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An apparatus comprises a processing device configured to obtain workload traces characterizing resource execution metrics associated with workloads executing in an information technology infrastructure and to generate sets of workload causality profiles characterizing relationships between the resource execution metrics, the sets of workload causality profiles comprising first and at least second sets of workload causality profiles generated using first and at least second different types of models. The processing device is also configured to determine sets of causal topics comprising recurring patterns of the relationships between different ones of the resource execution metrics found in each of the sets of workload causality profiles. The processing device is further configured to identify clusters of causal topics each comprising at least one causal topic that is part of the first and second sets of causal topics, and to perform diagnosis of issues affecting the workloads based on the identified clusters. ## Claims ### Claim 1 An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured: to execute a first set of a plurality of workloads on a set of storage systems of an information technology infrastructure, wherein executing the first set of the plurality of workloads produces a plurality of workload traces characterizing resource execution metrics associated with the execution of the first set of the plurality of workloads, wherein one of the resource execution metrics comprises an execution time of at least a subset of the first set of the plurality of workloads; responsive to an occurrence of one or more issues causing the execution time of the subset of the first set of the plurality of workloads to exceed an expected execution time, to at least one of (i) expand at least one storage system in the set of storage systems of the information technology infrastructure, and (ii) deploy one or more additional storage systems in the set of storage systems of the information technology infrastructure; and to execute a second set of the plurality of workloads on the set of storage systems in the information technology infrastructure; wherein the occurrence of the one or more issues is based at least in part on one or more clusters of causal topics that comprise at least one causal topic that is part of a first set of causal topics and a second set of causal topics, the first set of causal topics being based at least in part on a first set of recurring patterns of relationships between different ones of the resource execution metrics in a first set of workload causality profiles, the first set of workload causality profiles being based at least in part on a dependence measurement model and the plurality of workload traces, the second set of causal topics being based at least in part on a second set of recurring patterns of relationships between different ones of the resource execution metrics in a second set of workload causality profiles, and the second set of workload causality profiles being based at least in part on a causal model and the plurality of workload traces. ### Claim 2 (depends on 1) The apparatus of claim 1 wherein the resource execution metrics comprise two or more explanatory variable metrics and at least one response variable metric, and wherein the first and second sets of workload causality profiles characterize relationships between each of the two or more explanatory variable metrics and the at least one response variable metric. ### Claim 3 (depends on 2) The apparatus of claim 2 wherein the first and second sets of the plurality of workloads comprise backup operations executing in backup systems of the information technology infrastructure, and wherein the at least one response variable metric comprises the execution time. ### Claim 4 (depends on 3) The apparatus of claim 3 wherein the two or more explanatory variable metrics comprise two or more of: a backup operation type of the first set of the plurality of workloads executing in the backup systems of the information technology infrastructure; stream consumption on the backup systems of the information technology infrastructure; backup system model numbers of the backup systems of the information technology infrastructure; backup system processor usage of the backup systems of the information technology infrastructure; whether garbage collection processes are running on the backup systems of the information technology infrastructure; network issues associated with the backup systems of the information technology infrastructure; and amounts of writes to storage devices of the backup systems of the information technology infrastructure that exceed a designated latency threshold. ### Claim 5 (depends on 1) The apparatus of claim 1 wherein the dependence measurement model comprises a Schweizer-Wolff dependence measurement model. ### Claim 6 (depends on 1) The apparatus of claim 1 wherein the causal model comprises a Rubin causal model. ### Claim 7 (depends on 1) The apparatus of claim 1 wherein the first and second sets of causal topics are based at least in part on a topic model, wherein the workload causality profiles represent documents, the resource execution metrics represent words, and the causal topics comprise explanations for issues affecting the execution time of the first set of the plurality of workloads on the set of storage systems in the information technology infrastructure. ### Claim 8 (depends on 7) The apparatus of claim 7 wherein the topic model comprises a Latent Dirichlet Allocation (LDA) topic model. ### Claim 9 (depends on 7) The apparatus of claim 7 wherein a given one of the causal topics comprises probabilities of two or more of the resource execution metrics being a cause of one or more issues encountered during execution of a given one of the first set of the plurality of workloads executed on the set of storage systems of the information technology infrastructure. ### Claim 10 (depends on 1) The apparatus of claim 1 wherein the one or more clusters of causal topics are based at least in part on an ensemble learning algorithm. ### Claim 11 (depends on 10) The apparatus of claim 10 wherein the ensemble learning algorithm utilizes agglomerative hierarchical clustering. ### Claim 12 (depends on 1) The apparatus of claim 1 wherein the occurrence of the one or more issues affecting at least the subset of the first set of the plurality of workloads executed on the set of storage systems of the information technology infrastructure is based at least in part on explanations for the subset of the first set of the plurality of workloads violating one or more service level agreements. ### Claim 13 (depends on 1) The apparatus of claim 1 wherein at least one workload in the second set of the plurality of workloads comprises a re-execution of at least one workload in the first set of the plurality of workloads that failed during execution. ### Claim 14 (depends on 13) The apparatus of claim 13 wherein the at least one workload in the first set of the plurality of workloads failed during execution on a first subset of the set of storage systems the information technology infrastructure, and wherein the execution of said at least one workload in the second set of the plurality of workloads comprises execution on a second subset of the set of storage systems of the information technology infrastructure, the second set of the storage systems being different than the first set of the storage systems. ### Claim 15 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 execute a first set of a plurality of workloads on a set of storage systems of an information technology infrastructure, wherein executing the first set of the plurality of workloads produces a plurality of workload traces characterizing resource execution metrics associated with the execution of the first set of the plurality of workloads, wherein one of the resource execution metrics comprises an execution time of at least a subset of the first set of the plurality of workloads; responsive to an occurrence of one or more issues causing the execution time of the subset of the first set of the plurality of workloads to exceed an expected execution time, to at least one of (i) expand at least one storage system in the set of storage systems of the information technology infrastructure, and (ii) deploy one or more additional storage systems in the set of storage systems of the information technology infrastructure; and to execute a second set of the plurality of workloads on the set of storage systems of the information technology infrastructure; wherein the occurrence of the one or more issues is based at least in part on one or more clusters of causal topics that comprise at least one causal topic that is part of a first set of causal topics and a second set of causal topics, the first set of causal topics being based at least in part on a first set of recurring patterns of relationships between different ones of the resource execution metrics in a first set of workload causality profiles, the first set of workload causality profiles being based at least in part on a dependence measurement model and the plurality of workload traces, the second set of causal topics being based at least in part on a second set of recurring patterns of relationships between different ones of the resource execution metrics in a second set of workload causality profiles, and the second set of workload causality profiles being based at least in part on a causal model and the plurality of workload traces. ### Claim 16 (depends on 15) The computer program product of claim 15 wherein the resource execution metrics comprise two or more explanatory variable metrics and at least one response variable metric, and wherein the first and second sets of workload causality profiles characterize relationships between each of the two or more explanatory variable metrics and the at least one response variable metric. ### Claim 17 (depends on 16) The computer program product of claim 16 wherein the first and second sets of the plurality of workloads comprise backup operations executing in backup systems of the information technology infrastructure, and wherein the at least one response variable metric comprises the execution time. ### Claim 18 A method comprising: executing a first set of a plurality of workloads on a set of storage systems of an information technology infrastructure, wherein executing the first set of the plurality of workloads produces a plurality of workload traces characterizing resource execution metrics associated with the execution of the first set of the plurality of workloads, wherein one of the resource execution metrics comprises an execution time of at least a subset of the first set of the plurality of workloads; responsive to an occurrence of one or more issues causing the execution time of the subset of the first set of the plurality of workloads to exceed an expected execution time, at least one of (i) expanding at least one storage system in the set of storage systems of the information technology infrastructure, and (ii) deploying one or more additional storage systems in the set of storage systems of the information technology infrastructure; executing a second set of the plurality of workloads on the set of storage systems in the information technology infrastructure; wherein the occurrence of the one or more issues is based at least in part on one or more clusters of causal topics that comprise at least one causal topic that is part of a first set of causal topics and a second set of causal topics, the first set of causal topics being based at least in part on a first set of recurring patterns of relationships between different ones of the resource execution metrics in a first set of workload causality profiles, the first set of workload causality profiles being based at least in part on a dependence measurement model and the plurality of workload traces, the second set of causal topics being based at least in part on a second set of recurring patterns of relationships between different ones of the resource execution metrics in a second set of workload causality profiles, and the second set of workload causality profiles being based at least in part on a causal model and the plurality of workload traces; and wherein the method is performed by at least one processing device comprising a processor coupled to a memory. ### Claim 19 (depends on 18) The method of claim 18 wherein the resource execution metrics comprise two or more explanatory variable metrics and at least one response variable metric, and wherein the first and second sets of workload causality profiles characterize relationships between each of the two or more explanatory variable metrics and the at least one response variable metric. ### Claim 20 (depends on 19) The method of claim 19 wherein the first and second sets of the plurality of workloads comprise backup operations executing in backup systems of the information technology infrastructure, and wherein the at least one response variable metric comprises the execution time.