Ran this test against the MockHRegion, ran test for each duplicate row key percentage 100 times such that we reduce variance. Batch size is 25k rows. Note duplicate keys are shuffled throughout the batch randomly. Note: All times are in (ms) TRUNK (WITHOUT PATCH APPLIED) ************************************************************* STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.0 DescriptiveStatistics: n: 100 min: 30.0 max: 343.0 mean: 55.45000000000001 std dev: 32.44213963724377 median: 50.0 skewness: 7.236118305713237 kurtosis: 63.33779590624729 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.05 DescriptiveStatistics: n: 100 min: 59.0 max: 338.0 mean: 73.25 std dev: 34.20360161790515 median: 67.0 skewness: 6.6853135112154956 kurtosis: 46.728741851627994 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.1 DescriptiveStatistics: n: 100 min: 119.0 max: 192.0 mean: 127.77 std dev: 7.379818234327791 median: 127.0 skewness: 6.744632680348738 kurtosis: 58.63738016334204 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.15 DescriptiveStatistics: n: 100 min: 177.0 max: 263.0 mean: 191.68 std dev: 11.168027958634797 median: 191.5 skewness: 2.6318893750105254 kurtosis: 15.436994069891991 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.2 DescriptiveStatistics: n: 100 min: 253.0 max: 335.0 mean: 265.71 std dev: 9.340143619253551 median: 264.5 skewness: 4.198051163052898 kurtosis: 29.889075596629887 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.25 DescriptiveStatistics: n: 100 min: 331.0 max: 411.0 mean: 339.46 std dev: 8.761047859472042 median: 338.0 skewness: 5.692376348939405 kurtosis: 45.12420244165065 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.3 DescriptiveStatistics: n: 100 min: 406.0 max: 495.0 mean: 421.89 std dev: 10.999444430414558 median: 421.0 skewness: 2.972765470127306 kurtosis: 18.62809398602833 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.35 DescriptiveStatistics: n: 100 min: 497.0 max: 585.0 mean: 514.29 std dev: 11.492021826317844 median: 513.5 skewness: 2.300545596354651 kurtosis: 13.052939381707915 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.4 DescriptiveStatistics: n: 100 min: 598.0 max: 668.0 mean: 615.47 std dev: 11.575361329089898 median: 614.0 skewness: 1.444059148776078 kurtosis: 4.155668869564254 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.5 DescriptiveStatistics: n: 100 min: 774.0 max: 1109.0 mean: 817.24 std dev: 61.87264861362124 median: 801.0 skewness: 3.413785803572299 kurtosis: 11.843282315440375 ------------------------------------------------------------ TRUNK (WITH PROPOSED PATCH) ************************************************************* STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.0 DescriptiveStatistics: n: 100 min: 46.0 max: 435.0 mean: 99.84000000000003 std dev: 58.07113714830262 median: 85.0 skewness: 3.5101957571545754 kurtosis: 15.495906287057384 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.05 DescriptiveStatistics: n: 100 min: 36.0 max: 137.0 mean: 50.419999999999995 std dev: 15.183177825767865 median: 46.0 skewness: 2.896572549015482 kurtosis: 11.642380930897144 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.1 DescriptiveStatistics: n: 100 min: 33.0 max: 75.0 mean: 43.06 std dev: 7.22903796497085 median: 42.0 skewness: 1.4973283200273784 kurtosis: 3.737767525662687 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.15 DescriptiveStatistics: n: 100 min: 31.0 max: 67.0 mean: 44.0 std dev: 7.5250423670742626 median: 44.0 skewness: 0.44570032246238117 kurtosis: 0.19824326786656732 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.2 DescriptiveStatistics: n: 100 min: 30.0 max: 71.0 mean: 41.92 std dev: 7.603401684227734 median: 40.0 skewness: 0.72112690704603 kurtosis: 0.9172611669493356 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.25 DescriptiveStatistics: n: 100 min: 29.0 max: 86.0 mean: 39.60000000000001 std dev: 9.569522327460112 median: 38.0 skewness: 2.4280686322460627 kurtosis: 8.724688320494208 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.3 DescriptiveStatistics: n: 100 min: 25.0 max: 64.0 mean: 31.28 std dev: 5.896549239113752 median: 29.0 skewness: 2.5010592202399633 kurtosis: 9.743764790895183 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.35 DescriptiveStatistics: n: 100 min: 25.0 max: 58.0 mean: 31.92 std dev: 6.573100985505744 median: 30.0 skewness: 1.2492417050755957 kurtosis: 1.658245816657621 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.4 DescriptiveStatistics: n: 100 min: 22.0 max: 48.0 mean: 29.29 std dev: 4.969187889668519 median: 29.0 skewness: 0.8406586685347177 kurtosis: 0.6915096920342227 ------------------------------------------------------------ STATISTICS FOR DUPLICATE ROWKEY PERCENTAGE: 0.5 DescriptiveStatistics: n: 100 min: 18.0 max: 55.0 mean: 24.94 std dev: 5.348123431861335 median: 24.0 skewness: 2.4281059262051117 kurtosis: 10.411082912171173 ------------------------------------------------------------