Skip to content

feat: better prometheus buckets for batch_size, isl, and input length #847

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.

Sign up for GitHub

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

Open
michaelfeil wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat: better prometheus buckets for batch_size, isl, and input length #847

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
31 changes: 22 additions & 9 deletions router/src/prometheus.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,20 +22,33 @@ pub(crate) fn prometheus_builer(
duration_buckets.push(value);
}

// Input Length buckets
// Input Length buckets: half-steps up to 32k, then powers of 2
let input_length_matcher = Matcher::Full(String::from("te_request_input_length"));
let input_length_buckets: Vec<f64> = (0..20)
.map(|x| 2.0_f64.powi(x))
.filter(|x| (*x as usize) <= max_input_length)
.collect();
let input_length_buckets: Vec<f64> = [
1.0, 2.0, 4.0, 8.0, 16.0, 32.0, 64.0, 128.0, 256.0, 512.0, 768.0, 1024.0, 1536.0, 2048.0,
3072.0, 4096.0, 6144.0, 8192.0, 12288.0, 16384.0, 24576.0, 32768.0, 65536.0, 131072.0,
262144.0, 524288.0,
]
.into_iter()
.filter(|x| (*x as usize) <= max_input_length)
.collect();

// Batch size buckets
// Batch size buckets: 1-32 (all integers), then powers of 2
let batch_size_matcher = Matcher::Full(String::from("te_batch_next_size"));
let batch_size_buckets: Vec<f64> = (0..13).map(|x| 2.0_f64.powi(x)).collect();
let batch_size_buckets: Vec<f64> = (1..=32)
.map(|x| x as f64)
.chain((7..13).map(|x| 2.0_f64.powi(x)))
.collect();

// Batch tokens buckets
// Batch tokens buckets: half-steps up to 32k, then powers of 2
let batch_tokens_matcher = Matcher::Full(String::from("te_batch_next_tokens"));
let batch_tokens_buckets: Vec<f64> = (0..21).map(|x| 2.0_f64.powi(x)).collect();
let batch_tokens_buckets: Vec<f64> = [
1.0, 2.0, 4.0, 8.0, 16.0, 32.0, 64.0, 128.0, 256.0, 512.0, 768.0, 1024.0, 1536.0, 2048.0,
3072.0, 4096.0, 6144.0, 8192.0, 12288.0, 16384.0, 24576.0, 32768.0, 65536.0, 131072.0,
262144.0, 524288.0, 1048576.0,
]
.into_iter()
.collect();

// Prometheus handler
PrometheusBuilder::new()
Expand Down