Diagnosis of storage system component issues via data analytics

摘要

An improved technique involves mapping differences between current data collected from a component and previous data collected from the component to anomalous behavior of the component. A central diagnosis server collects previous state data such as disk or CPU utilization from components of various data storage systems it supports. The server may store this data, indexed by identifiers such as events linked to the state data, in a central database for later reference. The server then compares current state data being received from a particular data storage system to previous state data stored in the database. In some arrangements, the server selects previous state data based on matching event identifiers corresponding to the current state data and previous state data. The central diagnosis server then determines anomalous behavior based on the difference between current and previous state data.

主权项

1. A method of diagnosing a root cause of a problem in a component of a data storage system, the method comprising:
collecting current state data of the component in the data storage system over a current period of time, the current state data indicating a current state of the component at instants of time within the current period of time, the current state of the component at an instant of time being defined by current values of parameters of a set of parameters that describe the component; extracting, from storage, previous state data of the component in other data storage systems over a previous period of time, the previous state data indicating a previous state of the component at instants of time within the previous period of time, the previous state of the component at an instant of time being defined by previous values of the parameters; performing a comparison operation on the current state data and previous state data, the comparison operation generating a comparison result that indicates a difference between i) the current values of parameters over several of the instants of time within the current period of time and ii) the previous values of the parameters over several of the instants of time within the previous period of time; and mapping the difference to an anomalous behavior of the component over the current period of time, the anomalous behavior providing a diagnosis of the root cause of the problem.