[dag] broadcast nodes within window till all validators ack #11751

ibalajiarun · 2024-01-24T00:00:12Z

Description

Test Plan

ibalajiarun · 2024-01-24T00:00:22Z

Current dependencies on/for this PR:

main
- PR [dag] refactor/improve dag health monitoring #11362
  - PR [dag] stop voting based on pipeline latency #11407
    - PR [dag] concurrent dag store #11505
      - PR [dag] parallel message handler implementation #11685
        
        PR [dag] support parallel rb voting #11743
        
        PR [dag] broadcast nodes within window till all validators ack #11751 👈
        
        PR [dag] cache votes without re-counting #11808
        
        PR [dag] decouple payload 2/n #11918
        
        PR Balaji/dag forge #11433

This stack of pull requests is managed by Graphite.

trunk-io · 2024-01-24T00:00:37Z

⏱️ 5h 45m total CI duration on this PR

Job	Cumulative Duration	Recent Runs
rust-unit-tests	2h 25m	🟩 🟩 🟩 🟩
windows-build	2h 24m	🟩 🟩 🟩 🟩 🟩 (+1 more)
run-tests-main-branch	19m	🟩 🟩 🟥 🟥
check-dynamic-deps	12m	🟩 🟩 🟩 🟩 🟩 (+1 more)
rust-lints	10m	🟥 🟥 🟥
general-lints	8m	🟩 🟩 🟩
semgrep/ci	3m	🟩 🟩 🟩 🟩 🟩 (+2 more)
check	2m	⬜ 🟥 🟥 🟥 🟥 (+1 more)
file_change_determinator	41s	🟩 🟩 🟩 🟩
file_change_determinator	30s	⬜ 🟩 ⬜ 🟩 🟩
permission-check	27s	🟩 🟩 🟩 🟩 🟩
permission-check	16s	🟩 ⬜ 🟩 🟩 🟩 (+1 more)
permission-check	15s	🟩 🟩 🟩 🟩 🟩 (+1 more)
permission-check	12s	🟩 ⬜ 🟩 🟩 🟩 (+1 more)

🚨 3 jobs on the last run were significantly faster/slower than expected

Job	Duration	vs 7d avg
run-tests-main-branch	8m	4m
windows-build	30m	19m
rust-unit-tests	40m	31m

_{settings ⋅ feedback ⋅ docs ⋅ learn more about trunk.io}

sasha8 · 2024-01-27T09:04:43Z

consensus/src/dag/dag_driver.rs

@@ -93,7 +97,7 @@ impl DagDriver {
            payload_client,
            reliable_broadcast,
            time_service,
-            rb_abort_handle: Mutex::new(None),
+            rb_handles: Mutex::new(BoundedVecDeque::new(window_size_config as usize)),


Do we have to use the same window size here?

We could increase, but let's keep this for now.

zekun000 · 2024-02-08T01:20:55Z

consensus/src/dag/dag_driver.rs

-    fn drop(&mut self) {
-        if let Some((handle, _)) = self.rb_abort_handle.lock().as_ref() {
-            handle.abort()
+struct BoundedVecDeque<T> {


nit: probably move it to a util, and use it for the sliding window inside MetadataBackendAdapter too. also small unit tests would be good

github-actions · 2024-02-09T06:36:14Z

✅ Forge suite `compat` success on `aptos-node-v1.8.3` ==> `de78ba35c81508f7e778630eccfd0af8d50bd711`

Compatibility test results for aptos-node-v1.8.3 ==> de78ba35c81508f7e778630eccfd0af8d50bd711 (PR)
1. Check liveness of validators at old version: aptos-node-v1.8.3
compatibility::simple-validator-upgrade::liveness-check : committed: 4727 txn/s, latency: 6814 ms, (p50: 6500 ms, p90: 10400 ms, p99: 14500 ms), latency samples: 179660
2. Upgrading first Validator to new version: de78ba35c81508f7e778630eccfd0af8d50bd711
compatibility::simple-validator-upgrade::single-validator-upgrade : committed: 1805 txn/s, latency: 15641 ms, (p50: 19300 ms, p90: 22000 ms, p99: 22500 ms), latency samples: 92080
3. Upgrading rest of first batch to new version: de78ba35c81508f7e778630eccfd0af8d50bd711
compatibility::simple-validator-upgrade::half-validator-upgrade : committed: 1875 txn/s, latency: 15487 ms, (p50: 19100 ms, p90: 21900 ms, p99: 22200 ms), latency samples: 91920
4. upgrading second batch to new version: de78ba35c81508f7e778630eccfd0af8d50bd711
compatibility::simple-validator-upgrade::rest-validator-upgrade : committed: 3189 txn/s, latency: 9890 ms, (p50: 9800 ms, p90: 13800 ms, p99: 14700 ms), latency samples: 140340
5. check swarm health
Compatibility test for aptos-node-v1.8.3 ==> de78ba35c81508f7e778630eccfd0af8d50bd711 passed
Test Ok

github-actions · 2024-02-09T06:39:28Z

✅ Forge suite `realistic_env_max_load` success on `de78ba35c81508f7e778630eccfd0af8d50bd711`

two traffics test: inner traffic : committed: 7444 txn/s, latency: 5139 ms, (p50: 4800 ms, p90: 6400 ms, p99: 12200 ms), latency samples: 3215980
two traffics test : committed: 100 txn/s, latency: 2274 ms, (p50: 2100 ms, p90: 2500 ms, p99: 8600 ms), latency samples: 1840
Latency breakdown for phase 0: ["QsBatchToPos: max: 0.210, avg: 0.197", "QsPosToProposal: max: 0.183, avg: 0.163", "ConsensusProposalToOrdered: max: 0.575, avg: 0.532", "ConsensusOrderedToCommit: max: 0.502, avg: 0.458", "ConsensusProposalToCommit: max: 1.026, avg: 0.989"]
Max round gap was 1 [limit 4] at version 1467638. Max no progress secs was 4.769476 [limit 15] at version 1467638.
Test Ok

ibalajiarun requested review from zekun000 and sasha8 as code owners January 24, 2024 00:00

This was referenced Jan 24, 2024

[dag] refactor/improve dag health monitoring #11362

Merged

[dag] stop voting based on pipeline latency #11407

Merged

Balaji/dag forge #11433

Closed

[dag] concurrent dag store #11505

Merged

[dag] parallel message handler implementation #11685

Merged

ibalajiarun mentioned this pull request Jan 24, 2024

[dag] support parallel rb voting #11743

Merged

ibalajiarun force-pushed the balaji/concurrent-rb-handler branch from 2449040 to 89ba377 Compare January 24, 2024 00:43

ibalajiarun force-pushed the balaji/extended-rb branch from 10f4f30 to a472f03 Compare January 24, 2024 00:43

ibalajiarun force-pushed the balaji/concurrent-rb-handler branch from 89ba377 to d16b70c Compare January 24, 2024 15:43

ibalajiarun force-pushed the balaji/extended-rb branch from a472f03 to ee925d8 Compare January 24, 2024 15:43

ibalajiarun force-pushed the balaji/concurrent-rb-handler branch from d16b70c to b26c1f4 Compare January 25, 2024 22:00

ibalajiarun requested review from JoshLind and gregnazario as code owners January 25, 2024 22:00

ibalajiarun force-pushed the balaji/extended-rb branch from ee925d8 to 1b0b98e Compare January 25, 2024 22:00

sasha8 approved these changes Jan 27, 2024

View reviewed changes

ibalajiarun mentioned this pull request Jan 28, 2024

[dag] cache votes without re-counting #11808

Merged

ibalajiarun mentioned this pull request Feb 6, 2024

[dag] decouple payload 2/n #11918

Closed

zekun000 approved these changes Feb 8, 2024

View reviewed changes

ibalajiarun force-pushed the balaji/concurrent-rb-handler branch from b26c1f4 to 373112d Compare February 8, 2024 23:31

ibalajiarun force-pushed the balaji/extended-rb branch from ba071cb to 1deb748 Compare February 8, 2024 23:31

ibalajiarun force-pushed the balaji/concurrent-rb-handler branch from 373112d to 3f6f138 Compare February 9, 2024 03:12

Base automatically changed from balaji/concurrent-rb-handler to main February 9, 2024 04:07

ibalajiarun added 3 commits February 8, 2024 20:09

[dag] broadcast nodes within window till all validators ack

44a8f58

[dag] use DropGuard to drop handles

634e70a

move BoundedVecDeque to aptos-collections

012db41

ibalajiarun force-pushed the balaji/extended-rb branch from 1deb748 to 012db41 Compare February 9, 2024 05:01

use BoundedVecDeque in MetadataAdapter

e8c3ad1

ibalajiarun force-pushed the balaji/extended-rb branch from 36ce28f to e8c3ad1 Compare February 9, 2024 05:54

ibalajiarun enabled auto-merge (squash) February 9, 2024 05:54

lint

de78ba3

This comment has been minimized.

Sign in to view

ibalajiarun merged commit 2b3cce4 into main Feb 9, 2024
55 checks passed

ibalajiarun deleted the balaji/extended-rb branch February 9, 2024 06:41

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[dag] broadcast nodes within window till all validators ack #11751

[dag] broadcast nodes within window till all validators ack #11751

ibalajiarun commented Jan 24, 2024

ibalajiarun commented Jan 24, 2024 •

edited

Loading

trunk-io bot commented Jan 24, 2024 •

edited

Loading

sasha8 Jan 27, 2024

ibalajiarun Feb 9, 2024

zekun000 Feb 8, 2024

This comment has been minimized.

This comment has been minimized.

github-actions bot commented Feb 9, 2024

github-actions bot commented Feb 9, 2024

[dag] broadcast nodes within window till all validators ack #11751

[dag] broadcast nodes within window till all validators ack #11751

Conversation

ibalajiarun commented Jan 24, 2024

Description

Test Plan

ibalajiarun commented Jan 24, 2024 • edited Loading

trunk-io bot commented Jan 24, 2024 • edited Loading

sasha8 Jan 27, 2024

Choose a reason for hiding this comment

ibalajiarun Feb 9, 2024

Choose a reason for hiding this comment

zekun000 Feb 8, 2024

Choose a reason for hiding this comment

This comment has been minimized.

This comment has been minimized.

github-actions bot commented Feb 9, 2024

✅ Forge suite compat success on aptos-node-v1.8.3 ==> de78ba35c81508f7e778630eccfd0af8d50bd711

github-actions bot commented Feb 9, 2024

✅ Forge suite realistic_env_max_load success on de78ba35c81508f7e778630eccfd0af8d50bd711

ibalajiarun commented Jan 24, 2024 •

edited

Loading

trunk-io bot commented Jan 24, 2024 •

edited

Loading

✅ Forge suite `compat` success on `aptos-node-v1.8.3` ==> `de78ba35c81508f7e778630eccfd0af8d50bd711`

✅ Forge suite `realistic_env_max_load` success on `de78ba35c81508f7e778630eccfd0af8d50bd711`