rluvaton commented on PR #7996: URL: https://github.com/apache/arrow-rs/pull/7996#issuecomment-3124543776
Run the benchmark against this and: - #8008 these are the results locally: it shows 16 to 26 times faster ``` group add-benchmark-encoding-sliced-list perf-only-encode-actual-list-values ----- ---------------------------------- ----------------------------------- append_rows 10 large_list(0) of u64(0) 1.00 342.2±10.73ns ? ?/sec 1.17 400.1±13.97ns ? ?/sec append_rows 10 list(0) of u64(0) 1.00 404.6±16.02ns ? ?/sec 1.10 444.1±26.33ns ? ?/sec append_rows 4096 large_list(0) of u64(0) 1.00 72.9±3.35µs ? ?/sec 1.02 74.7±2.26µs ? ?/sec append_rows 4096 large_list(0) sliced to 10 of u64(0) 26.98 14.8±2.63µs ? ?/sec 1.00 548.9±14.61ns ? ?/sec append_rows 4096 list(0) of u64(0) 1.00 71.9±3.32µs ? ?/sec 1.02 73.5±3.44µs ? ?/sec append_rows 4096 list(0) sliced to 10 of u64(0) 22.10 13.8±3.40µs ? ?/sec 1.00 626.4±14.46ns ? ?/sec convert_columns 10 large_list(0) of u64(0) 1.00 638.2±18.00ns ? ?/sec 1.03 658.0±22.13ns ? ?/sec convert_columns 10 list(0) of u64(0) 1.00 663.0±17.46ns ? ?/sec 1.06 703.9±32.63ns ? ?/sec convert_columns 4096 large_list(0) of u64(0) 1.00 77.5±2.68µs ? ?/sec 1.01 78.1±4.30µs ? ?/sec convert_columns 4096 large_list(0) sliced to 10 of u64(0) 17.97 14.9±3.14µs ? ?/sec 1.00 830.4±23.24ns ? ?/sec convert_columns 4096 list(0) of u64(0) 1.00 70.3±1.55µs ? ?/sec 1.05 73.9±2.49µs ? ?/sec convert_columns 4096 list(0) sliced to 10 of u64(0) 16.38 14.3±2.74µs ? ?/sec 1.00 871.2±22.19ns ? ?/sec convert_columns_prepared 10 large_list(0) of u64(0) 1.00 422.5±8.89ns ? ?/sec 1.09 462.1±10.78ns ? ?/sec convert_columns_prepared 10 list(0) of u64(0) 1.00 466.6±16.83ns ? ?/sec 1.10 512.7±21.83ns ? ?/sec convert_columns_prepared 4096 large_list(0) of u64(0) 1.00 77.0±3.19µs ? ?/sec 1.02 78.4±4.27µs ? ?/sec convert_columns_prepared 4096 large_list(0) sliced to 10 of u64(0) 20.54 12.8±3.06µs ? ?/sec 1.00 622.3±12.86ns ? ?/sec convert_columns_prepared 4096 list(0) of u64(0) 1.00 69.5±2.28µs ? ?/sec 1.05 73.2±1.54µs ? ?/sec convert_columns_prepared 4096 list(0) sliced to 10 of u64(0) 19.09 12.9±2.62µs ? ?/sec 1.00 673.8±13.70ns ? ?/sec convert_rows 10 large_list(0) of u64(0) 1.00 1041.9±13.96ns ? ?/sec 1.01 1050.4±21.63ns ? ?/sec convert_rows 10 list(0) of u64(0) 1.00 1077.8±25.09ns ? ?/sec 1.03 1105.9±45.31ns ? ?/sec convert_rows 4096 large_list(0) of u64(0) 1.01 153.2±6.28µs ? ?/sec 1.00 152.1±4.65µs ? ?/sec convert_rows 4096 large_list(0) sliced to 10 of u64(0) 1.00 1345.3±24.29ns ? ?/sec 1.01 1361.6±23.10ns ? ?/sec convert_rows 4096 list(0) of u64(0) 1.00 145.8±6.87µs ? ?/sec 1.02 148.7±7.07µs ? ?/sec convert_rows 4096 list(0) sliced to 10 of u64(0) 1.00 1432.5±27.76ns ? ?/sec 1.01 1446.0±37.15ns ? ?/sec ``` -- 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]
