Implementing failover processes between storage stamps

摘要

Embodiments of the present invention relate to invoking and managing a failover of a storage account between partitions within a distributed computing environment, where each partition represents a key range of data for the storage account. The partitions affected by the failover include source partitions hosted on a primary storage stamp and destination partitions hosted on a secondary storage stamp, where the storage account's data is being actively replicated from the primary to the secondary storage stamp. Upon receiving a manual or automatic indication to perform the failover, configuring the source partitions to independently perform flush-send operations (e.g., distributing pending messages as a group) and then configuring the destination partitions to independently perform flush-replay operations (e.g., aggressively replaying currently pending transactions). Upon completing the flush-replay operations, designating the secondary storage stamp as a new primary storage stamp such that live traffic is directed to the new primary storage stamp.

主权项

1. A computer-implemented method in a distributed environment utilizing a processor and memory for invoking a clean failover of a storage account residing on storage stamps within the distributed computing environment, the method comprising:
providing a primary storage stamp that includes one or more source partitions configured for accepting live traffic and for sending messages, to one or more destination partitions in order to advance replication to a secondary storage stamp, wherein messages comprise transactions; providing the secondary storage stamp that includes the one or more destination partitions configured to replay the transactions of the messages; performing a failover between the primary and the secondary storage stamp for a subset of the data on the primary stamp, wherein the failover is an anticipated transition from the primary storage stamp to the secondary storage stamp; requesting that the one or more source partitions attempt to independently carry out a flush-send operation, wherein the flush-send operation involves distributing pending messages at the primary storage stamp to the one or more destination partitions as a group, wherein pending messages comprise logged messages at primary storage stamp; and reconfiguring the one or more destination partitions to independently carry out a flush-replay operation, wherein the flush-replay operation involves aggressively replaying the transactions of at least the pending messages received at the one or more destination partitions.