Published on Dec 27, 2022
Google Cloud recently announced the availability of Autoclass for Cloud Storage, an option that automates the lifecycle management of objects based on their last access time. This new feature simplifies cost optimization for workloads with unpredictable and unknown access patterns.
The Autoclass feature automates the transition of objects in a bucket according to their access patterns, supporting standard storage classes, nearline storage classes, coldline storage classes, and archive storage classes. New objects are managed by the new service, which starts them in standard storage and transitions them to colder storage classes after 30, 90, and 365 days if they are not accessed.
In exchange for a management fee, the service helps achieve price predictability by removing retrieval surcharges associated with colder storage classes; it is recommended if data in a bucket has a variety of access frequencies, or if the access patterns for the data are unknown or unpredictable. It is not recommended, however, if the data is compatible with a specific storage class, for example, backups.
The service can be monitored using the metrics autoclass/transition_operation_count, which indicates the number of storage class transitions automatically initiated, and autoclass/transitioned_bytes_count, which indicates the total number of bytes that have been transitioned.
Among the providers offering automatic data migration across storage classes is AWS, which introduced S3 Intelligent-Tiering in 2018.
Buckets with Autoclass are charged a monthly management fee of 0.0025 USD for every 1000 objects stored, with no retrieval fees or early deletion fees.
Currently, Autoclass must be enabled at the time of bucket creation via the CLI, API, or Cloud Console. The support for existing buckets is expected to be provided later in 2023. In the event that the feature is disabled, all the objects in the bucket will remain in their current storage class.
