Regardless of the redundancy policy adopted by the volume or whether the write request is eventually processed by the cache tier or capacity tier, the storage engine synchronously generates a data log when handling write requests. In most cases, this log contains the complete update details. Only when both the updates to the logs and the data itself have been persistently written, will the write request be marked as successful and returned to the client.
Although synchronously writing logs incurs some performance overhead, this approach ensures data integrity and validity through a log verification mechanism, especially in handling sudden failures such as power outages.
Data logs themselves do not need to occupy storage space long-term. Under normal circumstances, the storage engine releases the log-occupied space and reuses it after confirming the secure update of the data area, ensuring efficient utilization of storage resources.
