[FLINK-3257] Add Exactly-Once Processing Guarantees in Iterative DataStream Jobs

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Abandoned
• Component/s: API / DataStream
• Labels: pull-request-available stale-assigned
• Assignee: Unassigned
• Reporter: Paris Carbone
• Created: 18/Jan/16 19:11
• Updated: 20/Apr/21 08:24
• Resolved: 20/Apr/21 08:24

Description

The current snapshotting algorithm cannot support cycles in the execution graph. An alternative scheme can potentially include records in-transit through the back-edges of a cyclic execution graph (ABS [1]) to achieve the same guarantees.

One straightforward implementation of ABS for cyclic graphs can work as follows along the lines:

1) Upon triggering a barrier in an IterationHead from the TaskManager start block output and start upstream backup of all records forwarded from the respective IterationSink.

2) The IterationSink should eventually forward the current snapshotting epoch barrier to the IterationSource.

3) Upon receiving a barrier from the IterationSink, the IterationSource should finalize the snapshot, unblock its output and emit all records in-transit in FIFO order and continue the usual execution.
–
Upon restart the IterationSource should emit all records from the injected snapshot first and then continue its usual execution.

Several optimisations and slight variations can be potentially achieved but this can be the initial implementation take.

[1] http://arxiv.org/abs/1506.08603

Issue Links

• relates to: FLINK-9552 NPE in SpanningRecordSerializer during checkpoint with iterations
• links to: GitHub Pull Request #1668
• links to: GitHub Pull Request #1668
• mentioned in: Page Loading...