...

...

...

...

...

...

...

Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).

...

Compatibility, Deprecation, and Migration Plan

The goal of this FLIP is to provide a minimal viable product (MVP) that helps users with high and frequent backpressure. Unaligned checkpoints in MVP-stage will be disabled by default and can be enabled with a feature toggleinitially be an optional feature. After collecting experience and implementing all necessary extensions, unaligned checkpoint will probably be enabled by default for exactly once.

For compatibility, in documentation, clearly state that users of the checkpoint API can only use unaligned checkpoints if they do not require a consistent state across all operators.

In the following release, we aim to make unaligned checkpoints the default behavior and add the following enhancements:

...

...

Known Limitations

• State size increase
• Up to a couple of GB per task (especially painful on IO bound clusters)
• Depends on the actual policy used (probably UNALIGNED_WITH_MAX_INFLIGHT_DATA  is the more plausible default)
• Longer and heavier recovery depending on the increased state size
• Can potentially trigger death spiral after a first failure
• More up-to-date checkpoints will most likely still be an improvement about the current checkpoint behavior during backpressure
For the MVP
• no re-scaling (need to use savepoint)
  
• max in-flight checkpoints = 1

Test Plan

• Correctness tests with induced failures
• Compare checkpoints times under backpressure with current state
• Compare throughput under backpressure with current state
• Compare progress under backpressure with frequently induced failures
• Compare throughput with no checkpointing

Roadmap

There are two primary goals. First, we want to provide a minimal viable product (MVP) that breaks up the vicious cycle of overload and instabilities of a cluster with slow checkpoints and more accumulated load resulting from recoveries from outdated checkpoints. Second, we aim for the full FLIP implementation that will among other things allow rescaling from unaligned checkpoints to directly counter overload.

In particular, the following improvements will be achieved through full release over MVP:

• Re-scaling on unaligned checkpoints (need to use savepoint in MVP)
  
• Incremental checkpointing to not write the same buffer multiple times (when the job is backpressured)
• Advanced triggers for unaligned checkpoints such as timeouts on alignment or meeting the maximum threshold of checkpoint sizes
• Support for concurrent checkpoints
• Incremental loading and processing of state
• No additional memory to load channel state: ideally, existing network buffers should be reused
• Reduced number of files: single file could be reused for several checkpoints

We aim to make unaligned checkpoints the default behavior after the full implementation.