Skip to content

Comments

feat: only subscribe to sampled subnets#7860

Closed
twoeths wants to merge 1 commit intopeerDAS_devnet_7from
te/peerDAS_do_not_subscribe_to_all_subnets_2
Closed

feat: only subscribe to sampled subnets#7860
twoeths wants to merge 1 commit intopeerDAS_devnet_7from
te/peerDAS_do_not_subscribe_to_all_subnets_2

Conversation

@twoeths
Copy link
Contributor

@twoeths twoeths commented May 22, 2025

Motivation

  • right now we always subscribe to all column subnets which is not necessary because we only need columns that we custody/sample
  • that's why I see 128 logs like this per slot, and this could degrade performance. We only need 8 per slot
 debug: Received gossip dataColumn slot=198378, root=0x9945…49af, curentSlot=198378, peerId=16Uiu2HAm2J4ZRFnf7qWvu8VBRFT66E5XFWnRTwYTahB2oRJom5ji, delaySec=1.5910000801086426, gossipIndex=113, columnIndex=113, pending=data_column, haveColumns=31, expectedColumns=8, recvToValLatency=0.0009999275207519531, recvToValidation=0.0009999275207519531, validationTime=0

Description

  • only subscribe to custody/sampling subnets

matthewkeil pushed a commit that referenced this pull request Jul 15, 2025
**Motivation**

- this is a precondition for #7860
- when publishing a block, we need to ensure we have enough peers on all
column subnets so that we can publish all blobs data to the network
- this is not a concern for a node without validator but since this is
very easy to achieve, we don't need to care this case

**Description**

- track `samplingGroups` on PeerData
- for sampling groups, ensure at least 6 per subnets
- for non-sampling groups, ensure at least 4 per subnets

**Metrics on fusaka-devnet-2**
we have 25-35 peers on all subnets, so this is more than enough
<img width="1537" height="619" alt="Screenshot 2025-07-14 at 11 13 01"
src="https://github.com/user-attachments/assets/34c55997-e437-4c53-a829-a9c202a58d1d"
/>

---------

Co-authored-by: Tuyen Nguyen <twoeths@users.noreply.github.com>
@wemeetagain
Copy link
Member

@nflaig "we should fix this soonish"

@twoeths
Copy link
Contributor Author

twoeths commented Aug 12, 2025

closing in favor of #8181

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants