Continuous lock-minimal checkpointing and recovery with a distributed log-based datastore

A computer-implemented method comprising:

• obtaining, by a node in a cluster of nodes utilizing a log-based datastore, checkpointed data from a durable datastore, wherein the checkpointed data corresponds to local state derived from a sequence of updates to the log-based datastore;
• updating, by the node, a local data structure based on the obtained checkpointed data;
• attaching, by the node, to the log-based datastore at a specific position identified based on the checkpointed data;
• and
• replaying one or more updates, obtained from the log-based datastore via the attachment, to further update the local data structure to be current.

A system comprising:

• a first one or more electronic devices to implement a log-based datastore in a multi-tenant provider network;
• and
• a second one or more electronic devices to implement a cluster of nodes in the multi-tenant provider network, wherein a first node of the cluster includes instructions that upon execution cause the first node to:
• obtain checkpointed data from a durable datastore, wherein the checkpointed data corresponds to local state derived from a sequence of updates to the log-based datastore;
• update a local data structure based on the obtained checkpointed data;
• attach to the log-based datastore at a specific position based on use of the checkpointed data;
• and
• replay one or more updates, obtained from the log-based datastore via the attachment, to further update the local data structure to be current.