[data] anchored and jittered schedules #919
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
fwbrasil
force-pushed
the
schedule-improvements
branch
from
eab31e5 to
d612420
Compare
hearnadam
approved these changes
Dec 12, 2024
| effect.delayed(delay).andThen(Loop.continue(nextSchedule)) | ||
| }.getOrElse { | ||
| effect.map(Loop.done) | ||
| Clock.now.map { now => |
There was a problem hiding this comment.
Is it useful to fetch the clock once outside the loop to avoid pulling from the local multiple times?
| case Absent => Loop.done(state) | ||
| case Present((duration, nextSchedule)) => | ||
| clock.sleep(duration).map(_.use(_ => f(state).map(Loop.continue(_, nextSchedule)))) | ||
| Clock.now.map { now => |
There was a problem hiding this comment.
can we use lowercase clock?
| case Present((duration, nextSchedule)) => | ||
| val nextExecution = lastExecution + duration | ||
| clock.sleep(duration).map(_.use(_ => f(state).map(Loop.continue(nextExecution, _, nextSchedule)))) | ||
| Clock.now.map { now => |
Comment on lines
354
to
+356
| for | ||
| (d1, s1) <- a.next | ||
| (d2, s2) <- b.next | ||
| (d1, s1) <- a.next(now) | ||
| (d2, s2) <- b.next(now) |
There was a problem hiding this comment.
pre-existing: I think this should be a fold - otherwise you may end up discarding when one schedule produces a next.
There was a problem hiding this comment.
That was the intended behavior considering that the delay of a done schedule is essentially Instant.Max. The Min implementation does what you mentioned
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I'm working on STM optimizations and want to use a jittered schedule to reduce the impact of retry storms. Since we avoid side effects in
kyo-data, I'm passing the currentInstanttoschedule.nextand calculating the jitter deterministically via hashing. It isn't random but, given thatInstanthas nanosecond precision, it seems reasonable.We could pass a random number from
kyo-coreinstead but we'd need to eventually pass the instant anyway to implement more advanced schedules like the newSchedule.anchoredI've also added in this PR.