Abstract: Embodiments for deleting encryption keys in a data storage system by storing a current encryption key in a key table, the current key encrypting at least some data in one or more data containers of a filesystem of the data storage system. A key table maintains a starting container ID and an ending container ID for each container encrypted by the current encryption key, and a deleted container count counting a number of containers of the one or more data containers deleted from the file system. The process determines if the number of containers in the deleted container count equals a number of containers having data encrypted by the encryption key as determined by the starting container ID and ending container ID, and if so, marks the key for deletion in a garbage collection operation, which then deletes the key from the key table.

Abstract: Embodiments for rotating encryption keys in a sized-based process by defining a threshold value specifying a maximum amount of data to be encrypted by a single encryption key, determining whether or not data currently ingested by the data storage system exceeds the threshold value, and performing a key rotation operation to use a new key to encrypt incoming future data if it does exceed the threshold value. A time-based process performs key rotation from an old key to a new key in accordance with a periodic schedule, determines if the key rotation operation is successful in rotating to the new key from the old key, and if the key rotation operation is successful then performing a subsequent key rotation operation in accordance with the periodic schedule, or if not successful sending a user alert and automatically re-attempting the key rotation operation.

Abstract: One example method includes defining a physical namespace, determining a number of virtual namespaces, virtualizing the physical namespace by defining the virtual namespaces on the physical namespace, and generating a modified lookup key that is a function of a name of one of the virtual namespaces. The modified lookup key may be moved between virtual namespaces without requiring interruption of a backup or restore process, and without requiring an associated file system to be taken offline. Movement of the modified lookup key may be transparent to a user and may permit preservation of scripts that were in place prior to the move.

Abstract: Requests from file system services of a storage system are registered. Each file system service, when executed, utilizes one or more resources of the storage system. Each request includes information describing resource requirements required by a respective file system service. Resource utilization data of the resources are collected over a period of time. The resource utilization data includes an identification of a resource, a timestamp, and a measurement indicating a utilization level of the resource corresponding to the timestamp. A machine learning model is trained to predict utilization patterns of the resources. Execution of the file system services are scheduled based on the predicted utilization patterns. Monitoring is conducted during the execution of the file system services. Based on the monitoring a determination is made as to whether the machine learning model should be retrained.

Abstract: Requests from file system services of a storage system are registered. Each file system service, when executed, utilizes one or more resources of the storage system. Each request includes information describing resource requirements required by a respective file system service. Resource utilization data of the resources are collected over a period of time. The resource utilization data includes an identification of a resource, a timestamp, and a measurement indicating a utilization level of the resource corresponding to the timestamp. A machine learning model is trained to predict utilization patterns of the resources. Execution of the file system services are scheduled based on the predicted utilization patterns. Monitoring is conducted during the execution of the file system services. Based on the monitoring a determination is made as to whether the machine learning model should be retrained.