Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Reduce pipeline backpressure #14404

Merged
merged 4 commits into from
Aug 24, 2024
Merged
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions config/src/config/consensus_config.rs
Original file line number Diff line number Diff line change
Expand Up @@ -230,7 +230,7 @@ impl Default for ConsensusConfig {
min_blocks_to_activate: 4,
percentile: 0.5,
target_block_time_ms: 250,
min_block_time_ms_to_activate: 100,
min_block_time_ms_to_activate: 175,
// allow at least two spreading group from reordering in a single block, to utilize paralellism
min_calibrated_txns_per_block: 8,
}),
Expand All @@ -241,25 +241,25 @@ impl Default for ConsensusConfig {
// Block enters the pipeline after consensus orders it, and leaves the
// pipeline once quorum on execution result among validators has been reached
// (so-(badly)-called "commit certificate"), meaning 2f+1 validators have finished execution.
back_pressure_pipeline_latency_limit_ms: 800,
back_pressure_pipeline_latency_limit_ms: 1000,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Given our node is hitting 1s latency under expected load, I think this needs to be higher than 1s.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Increased this limit further.

max_sending_block_txns_after_filtering_override:
MAX_SENDING_BLOCK_TXNS_AFTER_FILTERING,
max_sending_block_bytes_override: 5 * 1024 * 1024,
backpressure_proposal_delay_ms: 100,
backpressure_proposal_delay_ms: 50,
},
PipelineBackpressureValues {
back_pressure_pipeline_latency_limit_ms: 1200,
back_pressure_pipeline_latency_limit_ms: 1300,
max_sending_block_txns_after_filtering_override:
MAX_SENDING_BLOCK_TXNS_AFTER_FILTERING,
max_sending_block_bytes_override: 5 * 1024 * 1024,
backpressure_proposal_delay_ms: 200,
backpressure_proposal_delay_ms: 100,
},
PipelineBackpressureValues {
back_pressure_pipeline_latency_limit_ms: 1600,
max_sending_block_txns_after_filtering_override:
MAX_SENDING_BLOCK_TXNS_AFTER_FILTERING,
max_sending_block_bytes_override: 5 * 1024 * 1024,
backpressure_proposal_delay_ms: 300,
backpressure_proposal_delay_ms: 200,
},
// with execution backpressure, only later start reducing block size
PipelineBackpressureValues {
Expand Down
Loading