US Patent Application 17726982. SHARED WORKLOAD ANOMALY DETECTION ACROSS MULTIPLE HOSTS BASED ON BARE-METAL RECOVERY AND SYSTEM-STATE RECOVERY DATA AND METADATA simplified abstract
SHARED WORKLOAD ANOMALY DETECTION ACROSS MULTIPLE HOSTS BASED ON BARE-METAL RECOVERY AND SYSTEM-STATE RECOVERY DATA AND METADATA
Organization Name
Inventor(s)
Shelesh Chopra of Bangalore (IN)
SHARED WORKLOAD ANOMALY DETECTION ACROSS MULTIPLE HOSTS BASED ON BARE-METAL RECOVERY AND SYSTEM-STATE RECOVERY DATA AND METADATA - A simplified explanation of the abstract
- This abstract for appeared for US patent application number 17726982 Titled 'SHARED WORKLOAD ANOMALY DETECTION ACROSS MULTIPLE HOSTS BASED ON BARE-METAL RECOVERY AND SYSTEM-STATE RECOVERY DATA AND METADATA'
Simplified Explanation
This abstract describes a method and system for detecting anomalies in shared workloads across multiple hosts. The system uses data and metadata from bare-metal recovery and system-state recovery to compare the behavior of workloads operating on different hosts. By comparing this metadata, patterns of data generation can be identified. Additionally, metadata generated by shared workload instances should be consistent across all hosts. If there are any discrepancies in the metadata generated by a shared workload instance on one or more hosts compared to a baseline instance, it could indicate unwanted anomalies. These anomalies can then be further analyzed and appropriate actions can be taken. The proposed embodiments aim to compare metadata from non-baseline shared workload instances with metadata from the baseline instance to detect and resolve anomalies related to shared workload behavior and functionality.
Original Abstract Submitted
A method and system for shared workload anomaly detection across multiple hosts based on bare-metal recovery and/or system-state recovery data and metadata. Since at least said bare-metal recovery and/or system-state recovery metadata has information about each workload operating on one or many host(s), comparing said metadata across any set of hosts could provide the pattern of data being generated there-across. Further, other metadata generated by any shared workload instance(s), operating on a set of hosts, should be consistent across the set of hosts. Should a shared workload instance on one or more hosts generate metadata not sync with metadata generated from a baseline shared workload instance, such discrepancies could be a sign of unwanted anomalies that may warrant further analyses, alerts, etc. Embodiments disclosed herein, accordingly, propose comparing metadata from non-baseline shared workload instances against metadata from the baseline shared workload instance, in order to detect and/or resolve anomalies concerning shared workload instance behavior and/or functionality.
