[murex971]

No description provided.

[codecov]

Codecov Report

❌ Patch coverage is 98.59551% with 5 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...tend/router/parser/rewrite/statement/projection.rs	84.21%	3 Missing ⚠️
pgdog/src/backend/pool/connection/buffer.rs	98.38%	1 Missing ⚠️
...og/src/backend/pool/connection/multi_shard/test.rs	98.63%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[github-actions]

Metal bench (arm64) failed. 1 of 148 measures outside bounds.

suite	throughput	cpu/txn	user cycles/txn	user instr/txn	kernel cycles/txn	ctx/txn
pooler		25.69 (-0.3% ± 0.9%)	25,732 (+1.0% ± 1.2%)	35,202 (-0.0% ± 0.02%)	37,121 (-0.7% ± 1.2%)
lb		26.63 (-0.3% ± 0.9%)	26,937 (+0.6% ± 0.8%)	36,083 (-0.0% ± 0.02%)	38,252 (-0.5% ± 1.3%)
sharding-w2		49.33 (+1.9%)	63,482 (+2.2%)	72,244 (-0.0%)	59,675 (+1.7%)	1.384 (-0.4%)
sharding-w8	159,017 (-0.7%)	46.08 (+0.1%)	66,233 (+0.5%)	56,057 (-0.0%)	52,692 (-0.6%)	0.1484 (+4.9%)

Counters per txn

suite	frontend stall	l1i refill	backend stall
pooler	11,147 (+2.3% ± 2.7%)	2,938 (+0.4% ± 0.4%)	3,064 (+0.4% ± 0.6%)
lb	11,781 (+1.0% ± 1.7%)	3,024 (+0.6% ± 0.4%)	3,341 (+1.4% ± 1.5%)
sharding-w2	34,443 (+4.0%)	5,016 (-0.4%)	7,363 (+2.0%)
sharding-w8	36,971 (+0.3%)	4,952 (-0.9%)	9,655 (+3.2%)

benchmark	measure	result	baseline	change	bound
simple_cold::join	latency	7,566	7,202	+5.06% ± 1.86%	+5%

Microbench: 134 of 135 measures within bounds.

• extended_cached: 14 of 15 benches slower, median +2.8% ± 1.0%
• extended_cold: 12 of 15 benches slower, median +1.8% ± 0.4%
• simple_cold: 12 of 15 benches slower, median +2.9% ± 0.6%

lb flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

pooler flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

sharding-w2 flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

sharding-w8 flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

Logs

lb, microbench, pooler, sharding-w2, sharding-w8, verdict.json

Compared c87fceb with 97fc159 on c7g.metal (on-demand).

[github-actions]

Bencher Report

Project	PgDog
Branch	having-execution
Testbed	Intel v1

🚨 1 Alert

Benchmark	Measure Units	View	Benchmark Result (Result Δ%)	Upper Boundary (Limit %)
lb-v3	cpu-per-txn Measure (units)	📈 plot 🚷 threshold 🚨 alert (🔔)	42.40 units (+1.90%) Baseline: 41.61 units	42.03 units (100.90%)

Click to view all benchmark results

Benchmark	Throughput	Benchmark Result operations / second (ops/s) x 1e3 (Result Δ%)	Lower Boundary operations / second (ops/s) x 1e3 (Limit %)	cpu-per-txn	Benchmark Result Measure (units) (Result Δ%)	Upper Boundary Measure (units) (Limit %)
lb-v3	📈 view plot 🚷 view threshold	15.99 ops/s x 1e3 (+0.00%) Baseline: 15.99 ops/s x 1e3	15.83 ops/s x 1e3 (99.00%)	📈 view plot 🚷 view threshold 🚨 view alert (🔔)	42.40 units (+1.90%) Baseline: 41.61 units	42.03 units (100.90%)

🐰 View full continuous benchmarking report in Bencher

[github-actions]

Bencher Report

Project	PgDog
Branch	having-execution
Testbed	Intel v1

Click to view all benchmark results

Benchmark	Throughput	Benchmark Result operations / second (ops/s) x 1e3 (Result Δ%)	Lower Boundary operations / second (ops/s) x 1e3 (Limit %)	cpu-per-txn	Benchmark Result Measure (units) (Result Δ%)	Upper Boundary Measure (units) (Limit %)
pooler-v3	📈 view plot 🚷 view threshold	16.99 ops/s x 1e3 (+0.00%) Baseline: 16.99 ops/s x 1e3	16.82 ops/s x 1e3 (99.00%)	📈 view plot 🚷 view threshold	40.91 units (+0.90%) Baseline: 40.55 units	40.95 units (99.90%)

🐰 View full continuous benchmarking report in Bencher

[murex971]

@jkaczman I think that failed bencher report is noise here

[jkaczman]

@jkaczman I think that failed bencher report is noise here

Probably. I'll be replacing Bencher this evening with x86 spot instances from AWS running within the metal bench to supplement the current arm64 Graviton data, which should fix the noise via the extra work our bench does to eliminate variance and provide 95% confidence intervals. I'll also fix that flaky ctx measure.

[jkaczman]

@murex971 can you rebase off main to get the new x86 bench in here?

[github-actions]

Metal bench (x86) failed. 3 of 148 measures outside bounds.

suite	throughput	cpu/txn	user cycles/txn	user instr/txn	kernel cycles/txn	ctx/txn
pooler		26.13 (-0.1% ± 0.2%)	20,453 (+0.0% ± 0.05%)	36,657 (-0.0% ± 0.09%)	33,323 (+0.0% ± 0.2%)
lb		26.51 (-0.2% ± 0.2%)	21,072 (+0.1% ± 0.1%)	37,609 (-0.0% ± 0.08%)	33,371 (-0.3% ± 0.4%)
sharding-w2		40.51 (+0.1%)	44,746 (-0.4%)	75,798 (-0.0%)	39,400 (+0.3%)	1.521 (-0.0%)
sharding-w8	175,380 (+0.4%)	38.47 (-0.3%)	49,066 (+0.0%)	58,978 (-0.0%)	41,588 (-0.7%)	0.1844 (-0.8%)

Counters per txn

suite	frontend stall	l1i refill	backend stall
pooler	3,454 (-0.5% ± 0.5%)	1,176 (+0.1% ± 0.5%)	42,876 (+1.1% ± 0.3%)
lb	3,629 (+2.0% ± 0.4%)	1,219 (+2.1% ± 0.4%)	43,904 (-0.4% ± 0.3%)
sharding-w2	6,201 (-2.0%)	1,711 (+1.0%)	120,063 (-0.0%)
sharding-w8	3,616 (+4.9%)	1,073 (+6.4%)	184,294 (-0.7%)

benchmark	measure	result	baseline	change	bound
extended_cached::join	latency	5,210	4,486	+16.15% ± 0.76%	+5%
extended_cold::join	latency	9,209	8,496	+8.39% ± 0.12%	+5%
simple_cold::join	latency	8,968	8,246	+8.77% ± 0.38%	+5%

Microbench: 132 of 135 measures within bounds.

• extended_cached: 12 of 15 benches slower, median +2.1% ± 0.7%

lb flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

microbench flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

pooler flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

sharding-w2 flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

sharding-w8 flamegraph vs 97fc159

Open interactive for hover, zoom and search.

Open interactive.

Logs

lb, microbench, pooler, sharding-w2, sharding-w8, verdict.json

Compared c87fceb with 97fc159 on m7i.metal-48xl (spot).