Use of Locks to Ensure Data Integrity Within and Across Agents
Objects are managed in a concurrent configuration so that multiple agents can read from and write to the same cache cluster and at times operate on the same set of objects.
Multiple threads in one agent can also behave in a similar manner, to enable concurrent RTCs.
Locking is one of the necessary costs of tuning inference agents for higher performance when concurrency features are used.
Locking is used to ensure the data you read is up-to-date, and to ensure that no other RTC is updating the same data concurrently.
Subtopics