Method for prediction of the duration of garbage collection for backup storage systems

摘要

Mechanisms for predicting a GC duration are described herein. In one embodiment, the mechanisms include receiving a first set of features determined based on current operating status and prior garbage collection (GC) statistics of a first storage system. In one embodiment, the mechanisms include predicting a GC duration of a first GC process being performed at the first storage system by applying a predictive model on the first set of features, wherein the predictive model was generated based on a second set of features received periodically from a plurality of storage systems.

主权项

1. A computer-implemented method, comprising:
receiving a first set of features determined based on current operating status and prior garbage collection (GC) statistics of a first storage system, wherein the first set of features include a number of files associated with the first storage system to be processed during a GC, wherein the first storage system is a backup storage system that backs up data from a plurality of client systems over a network; predicting a GC duration of a first GC process being performed at the first storage system by applying a predictive model on the first set of features of the first storage system, the GC duration representing an amount of time required to complete the first CC process at the first storage system, wherein the predictive model was generated based on a second set of features received periodically from a plurality of storage systems, wherein the second set of features include total GC durations of a plurality of GC operations performed in the plurality of storage systems; and performing a management action associated with the first storage system, in response to determining that the predicted GC duration of the first GC process exceeds a predetermined threshold.