Abstract
An orchestrator supports optimistic concurrency control across a distributed microservice system in which each service owns its own datastore. The orchestrator has a master broker at the source of a transaction and a broker agent on each participating service. The master broker indicates to the agents that a transaction uses optimistic concurrency. Each agent, on committing any transaction, records the changed data nodes into the context of any active optimistic transaction and merges them with data nodes already tracked, deferring abort because staging may later expand the transaction. Before commit, a detect phase invoked at every agent intersects the tracked data nodes with the patches being applied. A collision at any service is returned to the master broker with the colliding data nodes, preventing commit at all services. The master broker determines whether the changes are relevant or trivial and, if relevant, restarts the transaction on the updated datastore.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Anonymous, "Pre-Commit Conflict Detection and Coordinated Rebase for Optimistic Concurrency in Distributed Microservice Transactions", Technical Disclosure Commons, (September 23, 2026)
https://www.tdcommons.org/dpubs_series/11841