cardigan: update to 0.1.0-alpha.4 - #1421
Conversation
|
/benchmark-multiple --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
Benchmark ResultsFramework:
Full log |
|
/benchmark-multiple -f cardigan-json-tls,cardigan-h2,cardigan-h2c,cardigan-grpc,cardigan-grpc-tls --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
|
/benchmark-multiple -f cardigan-json-tls,cardigan-h2,cardigan-h2c,cardigan-grpc,cardigan-grpc-tls --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
|
@phjeltman ah sorry, those are acutal own entries - is there a reason we cannot merge them into a single framework entry (or maybe 2-3)? |
Benchmark ResultsFrameworks: 5 | Test: ✅
|
| Test | Conn | RPS | CPU | Mem | Δ RPS | Δ Mem |
|---|---|---|---|---|---|---|
| json-tls | 4096 | 1,763,949 | 6176.4% | 1.8GiB | +12.5% | -68.4% |
✅ cardigan-h2
| Test | Conn | RPS | CPU | Mem | Δ RPS | Δ Mem |
|---|---|---|---|---|---|---|
| baseline-h2 | 256 | 15,138,548 | 1644.1% | 1.2GiB | +4.0% | -76.0% |
| baseline-h2 | 1024 | 14,491,128 | 1658.8% | 2.3GiB | +4.4% | -60.3% |
| static-h2 | 256 | 1,189,184 | 6176.6% | 1.5GiB | +50.4% | -74.1% |
| static-h2 | 1024 | 1,046,258 | 6159.9% | 2.7GiB | +42.7% | -55.0% |
✅ cardigan-h2c
| Test | Conn | RPS | CPU | Mem | Δ RPS | Δ Mem |
|---|---|---|---|---|---|---|
| baseline-h2c | 256 | 15,224,612 | 1388.0% | 1.3GiB | +0.4% | -76.4% |
| baseline-h2c | 1024 | 14,860,276 | 1339.6% | 2.5GiB | -1.2% | -55.4% |
| baseline-h2c | 4096 | 14,053,901 | 1358.8% | 5.9GiB | +0.9% | +1.7% |
✅ cardigan-grpc
| Test | Conn | RPS | CPU | Mem | Δ RPS | Δ Mem |
|---|---|---|---|---|---|---|
| unary-grpc | 256 | 7,091,284 | 1406.6% | 2.0GiB | -1.2% | -62.3% |
| unary-grpc | 1024 | 6,901,980 | 1654.3% | 3.3GiB | -2.0% | -45.9% |
✅ cardigan-grpc-tls
| Test | Conn | RPS | CPU | Mem | Δ RPS | Δ Mem |
|---|---|---|---|---|---|---|
| unary-grpc-tls | 256 | 6,788,514 | 1484.9% | 2.1GiB | -2.6% | -63.2% |
| unary-grpc-tls | 1024 | 6,663,596 | 1763.3% | 3.8GiB | -0.5% | -43.3% |
|
@Kaliumhexacyanoferrat Yeah I should definitely merge them into one entry, this setup is a bit messy. I will do that in next PR. Thanks! |
|
/benchmark-multiple --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
|
|
|
/benchmark-multiple --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
|
/benchmark-multiple --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
|
So we obviously confused our benchmark workflow here … /benchmark -f cardigan --save |
|
👋 Benchmark request received. A collaborator will review and approve the run. |
Benchmark ResultsFramework:
Full log |
|
Hmm, consolidated entry looks meh in terms of memory. I will improve this a bit and be back with an alpha.5. |
Description
Bump Cardigan to
dev.cardigan:cardigan:0.1.0-alpha.4and update benchmark fixtures.