Limit preaggregating attestations - #5256
Merged
Merged
Conversation
Contributor
Performance Report鉁旓笍 no performance regression detected Full benchmark results
|
twoeths
marked this pull request as ready for review
March 10, 2023 10:05
dapplion
suggested changes
Mar 11, 2023
dapplion
left a comment
dapplion
left a comment
Contributor
There was a problem hiding this comment.
I'm not sure about the safety of this change in adverse conditions. This settings should be flags in case they need to be customized in the future
Member
Author
|
@dapplion are you worry about the
|
Contributor
|
Only |
twoeths
force-pushed
the
tuyen/attestationPool
branch
from
March 13, 2023 23:07
23b11ed to
bdefdef
Compare
dapplion
reviewed
Mar 14, 2023
|
|
||
| // validator gets SyncCommitteeContribution at 2/3 of slot, it's no use to preaggregate later than that time | ||
| if (this.clock.secFromSlot(slot) > this.cutOffSecFromSlot) { | ||
| throw new OpPoolError({code: OpPoolErrorCode.LATE_MESSAGE, slot}); |
Contributor
There was a problem hiding this comment.
How loud would be this error?
Member
Author
There was a problem hiding this comment.
we throw error here and gossipHandler would log error there.
since this is mainly for debugging purpose, not for end user so I changed log level to debug.
Contributor
There was a problem hiding this comment.
Why not return an insert outcome? Then the caller can decide to maybe log a debug if result is not add
wemeetagain
reviewed
Mar 23, 2023
dapplion
previously approved these changes
Mar 28, 2023
Co-authored-by: Cayman <caymannava@gmail.com>
twoeths
force-pushed
the
tuyen/attestationPool
branch
from
March 28, 2023 02:39
8580147 to
7eaa0d0
Compare
dapplion
approved these changes
Mar 29, 2023
wemeetagain
approved these changes
Mar 29, 2023
twoeths
added a commit
that referenced
this pull request
Mar 29, 2023
* Limit preaggregating attestations * preaggregateSlotDistance hidden cli param * Log debug if error adding SyncCommitteeMessage to pool * SyncCommitteeMessagePool: return instead of throw error * Add SyncCommitteeMesssage insertOutcome metric * Update prune() method header Co-authored-by: Cayman <caymannava@gmail.com> --------- Co-authored-by: Cayman <caymannava@gmail.com>
Member
|
馃帀 This PR is included in v1.8.0 馃帀 |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Motivation
As shown on a subscribe-all-subnets goerli node, preaggregating attestations takes 8% of cpu time, on mainnet it could be worse. I think some preaggregation are redundant because validators only get AggregatedAttestation at 2/3 of clock slot, including:
slot < clockSlotand> clockSlot - 3slot = clockSlotbut come to the pool at > 2/3 of slotI noticed that on mainnet node, the attestation job wait time is 7s in average at some points so they are not useful to preaggregate anymore
Description
part of #5247