adriangbot commented on PR #26062: URL: https://github.com/apache/datafusion/pull/26062#issuecomment-6006958595
🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/26062#issuecomment-6006040989) **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) Comparing neilc/perf-sqrt (c9a5496ab877cd81cab797de31023c89f5da5aa2) to 8248a57 (merge-base) [diff](https://github.com/apache/datafusion/compare/8248a579695b92b7ca94709eedd63a71689513ce..c9a5496ab877cd81cab797de31023c89f5da5aa2) <details><summary>Run configuration</summary> ```yaml run benchmark math_expressions ``` </details> <details><summary>CPU Details (lscpu)</summary> ``` Architecture: aarch64 CPU op-mode(s): 64-bit Byte Order: Little Endian CPU(s): 16 On-line CPU(s) list: 0-15 Vendor ID: ARM Model name: Neoverse-V2 Model: 1 Thread(s) per core: 1 Core(s) per cluster: 16 Socket(s): - Cluster(s): 1 Stepping: r0p1 BogoMIPS: 2000.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti L1d cache: 1 MiB (16 instances) L1i cache: 1 MiB (16 instances) L2 cache: 32 MiB (16 instances) L3 cache: 80 MiB (1 instance) NUMA node(s): 1 NUMA node0 CPU(s): 0-15 Vulnerability Gather data sampling: Not affected Vulnerability Indirect target selection: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Reg file data sampling: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; __user pointer sanitization Vulnerability Spectre v2: Mitigation; CSV2, BHB Vulnerability Srbds: Not affected Vulnerability Tsa: Not affected Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Not affected ``` </details> <details><summary>Details</summary> <p> ``` group HEAD neilc_perf-sqrt ----- ---- --------------- atan2 f32 array: 1024 1.00 9.8±0.01µs ? ?/sec 1.00 9.8±0.01µs ? ?/sec atan2 f32 array: 4096 1.00 38.7±0.09µs ? ?/sec 1.00 38.7±0.03µs ? ?/sec atan2 f32 array: 8192 1.00 78.4±0.04µs ? ?/sec 1.01 78.8±0.16µs ? ?/sec atan2 f32 scalar 1.00 158.3±0.32ns ? ?/sec 1.02 161.6±1.86ns ? ?/sec atan2 f64 array: 1024 1.01 9.7±0.01µs ? ?/sec 1.00 9.6±0.01µs ? ?/sec atan2 f64 array: 4096 1.00 38.0±0.06µs ? ?/sec 1.00 37.9±0.03µs ? ?/sec atan2 f64 array: 8192 1.00 75.9±0.30µs ? ?/sec 1.00 75.9±0.05µs ? ?/sec atan2 f64 scalar 1.00 146.8±0.96ns ? ?/sec 1.04 152.2±3.04ns ? ?/sec cot f32 array: 1024 1.00 7.1±0.01µs ? ?/sec 1.00 7.1±0.00µs ? ?/sec cot f32 array: 4096 1.00 27.0±0.02µs ? ?/sec 1.00 27.1±0.01µs ? ?/sec cot f32 array: 8192 1.00 57.9±0.11µs ? ?/sec 1.00 57.8±0.11µs ? ?/sec cot f32 scalar 1.00 136.9±0.63ns ? ?/sec 1.01 137.9±0.78ns ? ?/sec cot f64 array: 1024 1.00 6.3±0.01µs ? ?/sec 1.00 6.3±0.00µs ? ?/sec cot f64 array: 4096 1.00 23.9±0.01µs ? ?/sec 1.00 23.9±0.02µs ? ?/sec cot f64 array: 8192 1.00 54.1±0.18µs ? ?/sec 1.00 54.1±0.14µs ? ?/sec cot f64 scalar 1.00 125.6±0.93ns ? ?/sec 1.01 126.4±0.98ns ? ?/sec degrees/f32 1.00 642.0±2.04ns ? ?/sec degrees/f32 with nulls 1.00 657.6±3.09ns ? ?/sec degrees/f64 1.00 1057.9±4.56ns ? ?/sec degrees/f64 with nulls 1.00 1072.2±9.65ns ? ?/sec factorial_array 1.00 3.0±0.10µs ? ?/sec 1.00 3.0±0.10µs ? ?/sec factorial_scalar 1.00 161.7±2.30ns ? ?/sec 1.03 166.9±2.82ns ? ?/sec floor_ceil size=1024/ceil_f64_array 1.00 374.0±0.80ns ? ?/sec 1.01 377.2±1.26ns ? ?/sec floor_ceil size=1024/ceil_f64_scalar 1.00 126.7±0.09ns ? ?/sec 1.00 126.9±0.20ns ? ?/sec floor_ceil size=1024/floor_f64_array 1.02 381.7±0.28ns ? ?/sec 1.00 376.1±0.30ns ? ?/sec floor_ceil size=1024/floor_f64_scalar 1.00 127.4±0.11ns ? ?/sec 1.05 134.2±0.14ns ? ?/sec floor_ceil size=4096/ceil_f64_array 1.00 668.8±0.19ns ? ?/sec 1.01 677.3±0.38ns ? ?/sec floor_ceil size=4096/ceil_f64_scalar 1.00 126.6±0.06ns ? ?/sec 1.00 126.6±0.09ns ? ?/sec floor_ceil size=4096/floor_f64_array 1.00 676.0±0.75ns ? ?/sec 1.00 677.0±0.60ns ? ?/sec floor_ceil size=4096/floor_f64_scalar 1.01 127.2±0.04ns ? ?/sec 1.00 125.9±0.07ns ? ?/sec floor_ceil size=8192/ceil_f64_array 1.00 1081.0±2.52ns ? ?/sec 1.00 1081.1±1.48ns ? ?/sec floor_ceil size=8192/ceil_f64_scalar 1.00 126.7±0.15ns ? ?/sec 1.00 126.5±0.10ns ? ?/sec floor_ceil size=8192/floor_f64_array 1.01 1094.3±1.96ns ? ?/sec 1.00 1082.1±2.70ns ? ?/sec floor_ceil size=8192/floor_f64_scalar 1.00 126.8±0.03ns ? ?/sec 1.00 126.5±0.19ns ? ?/sec gcd array and scalar 1.00 1771.2±2.41µs ? ?/sec 1.00 1776.8±5.17µs ? ?/sec gcd both array 1.00 1416.2±2.94µs ? ?/sec 1.00 1419.5±2.81µs ? ?/sec gcd both scalar 1.00 195.9±3.74ns ? ?/sec 1.01 198.8±4.55ns ? ?/sec isnan f32 array: 1024 1.00 441.5±1.32ns ? ?/sec 1.00 442.7±0.96ns ? ?/sec isnan f32 array: 4096 1.00 1034.2±1.47ns ? ?/sec 1.00 1038.5±4.24ns ? ?/sec isnan f32 array: 8192 1.00 1834.7±2.30ns ? ?/sec 1.01 1845.1±3.21ns ? ?/sec isnan f64 array: 1024 1.00 403.7±1.32ns ? ?/sec 1.00 403.1±1.75ns ? ?/sec isnan f64 array: 4096 1.00 882.2±1.35ns ? ?/sec 1.00 880.8±1.09ns ? ?/sec isnan f64 array: 8192 1.00 1542.0±2.65ns ? ?/sec 1.01 1560.7±2.58ns ? ?/sec iszero f32 array: 1024 1.00 388.4±2.31ns ? ?/sec 1.01 391.2±3.48ns ? ?/sec iszero f32 array: 4096 1.00 900.2±2.27ns ? ?/sec 1.01 906.7±1.47ns ? ?/sec iszero f32 array: 8192 1.00 1584.6±2.18ns ? ?/sec 1.01 1598.1±2.69ns ? ?/sec iszero f32 scalar 1.02 107.8±0.50ns ? ?/sec 1.00 105.2±0.53ns ? ?/sec iszero f64 array: 1024 1.00 374.6±2.27ns ? ?/sec 1.01 377.7±2.34ns ? ?/sec iszero f64 array: 4096 1.00 854.3±2.27ns ? ?/sec 1.00 857.6±2.92ns ? ?/sec iszero f64 array: 8192 1.00 1511.0±2.33ns ? ?/sec 1.00 1512.0±3.00ns ? ?/sec iszero f64 scalar 1.00 98.6±0.55ns ? ?/sec 1.00 98.5±0.73ns ? ?/sec lcm both array 1.00 1518.1±3.93µs ? ?/sec 1.00 1512.1±1.97µs ? ?/sec nanvl/array_f32/1024 1.00 408.3±1.95ns ? ?/sec 1.01 412.1±1.52ns ? ?/sec nanvl/array_f32/4096 1.00 649.9±2.41ns ? ?/sec 1.04 675.0±3.88ns ? ?/sec nanvl/array_f32/8192 1.00 1008.6±4.18ns ? ?/sec 1.04 1054.0±2.36ns ? ?/sec nanvl/array_f64/1024 1.02 496.6±2.50ns ? ?/sec 1.00 485.3±1.86ns ? ?/sec nanvl/array_f64/4096 1.02 1032.7±4.85ns ? ?/sec 1.00 1014.5±3.06ns ? ?/sec nanvl/array_f64/8192 1.00 1778.6±3.27ns ? ?/sec 1.09 1935.5±3.71ns ? ?/sec nanvl/array_f64_both_nulls/1024 1.00 2.9±0.15µs ? ?/sec 1.00 2.9±0.15µs ? ?/sec nanvl/array_f64_both_nulls/4096 1.00 10.3±0.68µs ? ?/sec 1.00 10.3±0.72µs ? ?/sec nanvl/array_f64_both_nulls/8192 1.00 20.2±1.33µs ? ?/sec 1.00 20.2±1.51µs ? ?/sec nanvl/array_f64_x_nulls/1024 1.00 2.9±0.18µs ? ?/sec 1.00 2.9±0.18µs ? ?/sec nanvl/array_f64_x_nulls/4096 1.00 10.2±0.78µs ? ?/sec 1.01 10.3±0.77µs ? ?/sec nanvl/array_f64_x_nulls/8192 1.00 20.0±1.59µs ? ?/sec 1.02 20.4±1.42µs ? ?/sec nanvl/array_f64_y_nulls/1024 1.01 3.0±0.12µs ? ?/sec 1.00 3.0±0.14µs ? ?/sec nanvl/array_f64_y_nulls/4096 1.00 10.5±0.52µs ? ?/sec 1.06 11.2±0.25µs ? ?/sec nanvl/array_f64_y_nulls/8192 1.00 20.3±1.30µs ? ?/sec 1.08 22.0±0.54µs ? ?/sec nanvl/scalar_f32 1.00 131.3±1.40ns ? ?/sec 1.01 132.0±2.76ns ? ?/sec nanvl/scalar_f64 1.00 119.1±1.09ns ? ?/sec 1.01 120.3±3.15ns ? ?/sec power f64 array x f64 array, n=1024 1.00 8.8±0.02µs ? ?/sec 1.01 8.9±0.02µs ? ?/sec power f64 array x f64 array, n=8192 1.00 65.8±0.16µs ? ?/sec 1.00 65.9±0.18µs ? ?/sec power f64 array x f64 scalar, exp=0.5, n=1024 1.00 9.2±0.05µs ? ?/sec 1.00 9.3±0.05µs ? ?/sec power f64 array x f64 scalar, exp=0.5, n=8192 1.00 71.6±0.38µs ? ?/sec 1.00 71.7±0.39µs ? ?/sec power f64 array x f64 scalar, exp=2, n=1024 1.00 9.3±0.05µs ? ?/sec 1.00 9.3±0.05µs ? ?/sec power f64 array x f64 scalar, exp=2, n=8192 1.00 71.6±0.39µs ? ?/sec 1.00 71.7±0.39µs ? ?/sec random_1M_rows_batch_128 1.00 13.9±0.54ms ? ?/sec 1.00 14.0±0.55ms ? ?/sec random_1M_rows_batch_8192 1.00 12.4±0.57ms ? ?/sec 1.00 12.4±0.57ms ? ?/sec round size=1024/round_f32_array 1.00 604.1±1.67ns ? ?/sec 1.02 617.0±1.45ns ? ?/sec round size=1024/round_f32_scalar 1.00 184.6±1.22ns ? ?/sec 1.04 192.3±0.86ns ? ?/sec round size=1024/round_f64_array 1.00 1214.8±5.24ns ? ?/sec 1.01 1223.3±1.64ns ? ?/sec round size=1024/round_f64_scalar 1.00 186.9±0.88ns ? ?/sec 1.01 188.7±0.36ns ? ?/sec round size=4096/round_f32_array 1.00 1377.5±3.70ns ? ?/sec 1.01 1394.1±2.41ns ? ?/sec round size=4096/round_f32_scalar 1.00 185.5±1.39ns ? ?/sec 1.03 191.5±1.21ns ? ?/sec round size=4096/round_f64_array 1.00 3.8±0.00µs ? ?/sec 1.00 3.8±0.00µs ? ?/sec round size=4096/round_f64_scalar 1.00 184.2±1.30ns ? ?/sec 1.01 186.8±0.58ns ? ?/sec round size=8192/round_f32_array 1.00 2.4±0.00µs ? ?/sec 1.00 2.4±0.00µs ? ?/sec round size=8192/round_f32_scalar 1.00 185.4±1.39ns ? ?/sec 1.04 192.8±1.24ns ? ?/sec round size=8192/round_f64_array 1.00 7.2±0.00µs ? ?/sec 1.00 7.2±0.00µs ? ?/sec round size=8192/round_f64_scalar 1.00 183.2±1.27ns ? ?/sec 1.02 186.7±0.70ns ? ?/sec round_dense_f32/1024 1.00 585.0±1.47ns ? ?/sec 1.02 594.8±5.01ns ? ?/sec round_dense_f32/4096 1.00 1367.7±3.82ns ? ?/sec 1.00 1366.8±4.99ns ? ?/sec round_dense_f32/8192 1.00 2.4±0.00µs ? ?/sec 1.00 2.4±0.00µs ? ?/sec round_dense_f64/1024 1.00 1192.8±2.37ns ? ?/sec 1.01 1204.8±7.62ns ? ?/sec round_dense_f64/4096 1.00 3.8±0.00µs ? ?/sec 1.00 3.8±0.00µs ? ?/sec round_dense_f64/8192 1.00 7.2±0.00µs ? ?/sec 1.00 7.2±0.01µs ? ?/sec signum f32 array: 1024 1.00 402.6±0.97ns ? ?/sec 1.02 411.3±2.12ns ? ?/sec signum f32 array: 4096 1.00 873.5±1.09ns ? ?/sec 1.00 875.0±1.11ns ? ?/sec signum f32 array: 8192 1.00 1501.0±1.21ns ? ?/sec 1.00 1497.0±1.23ns ? ?/sec signum f32 scalar: 1024 1.05 115.9±6.56ns ? ?/sec 1.00 110.3±0.26ns ? ?/sec signum f32 scalar: 4096 1.00 108.8±0.21ns ? ?/sec 1.01 110.3±0.26ns ? ?/sec signum f32 scalar: 8192 1.00 108.9±0.36ns ? ?/sec 1.01 110.3±0.24ns ? ?/sec signum f64 array: 1024 1.00 1017.8±1.55ns ? ?/sec 1.01 1022.9±2.95ns ? ?/sec signum f64 array: 4096 1.00 3.3±0.00µs ? ?/sec 1.00 3.3±0.00µs ? ?/sec signum f64 array: 8192 1.00 6.4±0.00µs ? ?/sec 1.00 6.4±0.01µs ? ?/sec signum f64 scalar: 1024 1.00 100.8±0.90ns ? ?/sec 1.01 102.2±0.78ns ? ?/sec signum f64 scalar: 4096 1.06 107.9±2.95ns ? ?/sec 1.00 102.1±0.71ns ? ?/sec signum f64 scalar: 8192 1.00 100.9±0.90ns ? ?/sec 1.01 102.2±0.71ns ? ?/sec sqrt/f32 1.00 2.7±0.01µs ? ?/sec sqrt/f32 with nulls 1.00 2.7±0.00µs ? ?/sec sqrt/f64 1.00 9.9±0.01µs ? ?/sec sqrt/f64 with nulls 1.00 9.9±0.00µs ? ?/sec trunc f32 array: 1024 1.00 365.4±1.27ns ? ?/sec 1.00 366.9±1.11ns ? ?/sec trunc f32 array: 4096 1.00 675.6±2.85ns ? ?/sec 1.00 674.6±2.47ns ? ?/sec trunc f32 array: 8192 1.00 1100.6±1.11ns ? ?/sec 1.00 1105.7±1.80ns ? ?/sec trunc f32 precision array: 1024 1.00 502.6±0.96ns ? ?/sec 1.01 507.4±2.66ns ? ?/sec trunc f32 precision array: 4096 1.00 1287.4±5.30ns ? ?/sec 1.00 1288.0±3.09ns ? ?/sec trunc f32 precision array: 8192 1.00 2.3±0.00µs ? ?/sec 1.00 2.3±0.00µs ? ?/sec trunc f32 scalar 1.00 67.1±0.40ns ? ?/sec 1.00 66.9±0.34ns ? ?/sec trunc f64 array: 1024 1.00 423.4±2.06ns ? ?/sec 1.01 428.8±1.93ns ? ?/sec trunc f64 array: 4096 1.00 916.3±3.30ns ? ?/sec 1.01 923.8±2.07ns ? ?/sec trunc f64 array: 8192 1.00 1564.4±3.47ns ? ?/sec 1.00 1558.1±3.93ns ? ?/sec trunc f64 precision array: 1024 1.00 1089.1±1.73ns ? ?/sec 1.01 1096.0±2.70ns ? ?/sec trunc f64 precision array: 4096 1.00 3.6±0.00µs ? ?/sec 1.00 3.6±0.00µs ? ?/sec trunc f64 precision array: 8192 1.00 7.0±0.00µs ? ?/sec 1.00 7.0±0.00µs ? ?/sec trunc f64 scalar 1.01 64.0±0.55ns ? ?/sec 1.00 63.7±0.66ns ? ?/sec ``` </p> </details> <details><summary>Resource Usage</summary> **math_expressions — base (merge-base)** | Metric | Value | |--------|-------| | Wall time | 1390.3s | | Peak memory | 37.0 MiB | | Avg memory | 30.0 MiB | | CPU user | 1484.4s | | CPU sys | 1.3s | | Peak spill | 0 B | **math_expressions — branch** | Metric | Value | |--------|-------| | Wall time | 1475.3s | | Peak memory | 38.1 MiB | | Avg memory | 31.6 MiB | | CPU user | 1587.6s | | CPU sys | 1.4s | | Peak spill | 0 B | </details> --- [File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
