alexey-milovidov · Oct 12, 2026 · Oct 6, 2026 · Oct 12, 2026 · Oct 12, 2026 · Oct 12, 2026
diff --git a/dask/queries.sql b/dask/queries.sql
@@ -20,7 +20,7 @@ hits.groupby([hits['UserID'], hits['EventTime'].dt.minute, 'SearchPhrase']).size
 hits[hits['UserID'] == 435090932899640449]
 hits[hits['URL'].str.contains('google')].shape[0]
 hits[hits['URL'].str.contains('google') & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg(URL=('URL', 'min'), c=('SearchPhrase', 'size')).nlargest(10, 'c')
-hits[hits['Title'].str.contains('Google') & ~hits['URL'].str.contains('.google.') & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg(URL=('URL', 'min'), Title=('Title', 'min'), c=('SearchPhrase', 'size'), UserID=('UserID', nunique)).nlargest(10, 'c')
+hits[hits['Title'].str.contains('Google') & ~hits['URL'].str.contains('.google.', regex=False) & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg(URL=('URL', 'min'), Title=('Title', 'min'), c=('SearchPhrase', 'size'), UserID=('UserID', nunique)).nlargest(10, 'c')
 hits[hits['URL'].str.contains('google')].nsmallest(10, 'EventTime')
 hits[hits['SearchPhrase'] != ''].nsmallest(10, 'EventTime').compute()[['SearchPhrase']]
 hits[hits['SearchPhrase'] != ''][['SearchPhrase']].sort_values('SearchPhrase').head(10, npartitions=-1)

diff --git a/dask/results/20261012/c6a.metal.json b/dask/results/20261012/c6a.metal.json
@@ -0,0 +1,60 @@
+{
+      "system": "Dask (DataFrame)",
+      "date": "2026-10-12",
+      "machine": "c6a.metal",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","column-oriented"],
+      "load_time": 6,
+      "data_size": 174168110180,
+      "concurrent_qps": 0.078,
+      "concurrent_error_ratio": 0.175,
+      "result": [
+        [64.395, 7.368, 8.014],
+        [58.197, 0.845, 0.893],
+        [57.605, 0.334, 0.25],
+        [57.294, 0.284, 0.18],
+        [58.564, 1.129, 1.386],
+        [59.54, 2.226, 2.147],
+        [57.556, 0.287, 0.174],
+        [58.225, 0.736, 0.58],
+        [59.688, 2.377, 2.337],
+        [null, null, null],
+        [67.346, 10.024, 10.865],
+        [68.265, 11.004, 11.138],
+        [61.693, 4.038, 4.134],
+        [68.193, 10.958, 9.915],
+        [116.339, 51.11, 49.199],
+        [59.034, 1.42, 1.697],
+        [143.261, 96.192, 93.699],
+        [64.308, 6.721, 7.137],
+        [143.042, 83.062, 85.969],
+        [57.742, 0.593, 0.495],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [58.371, 1.04, 1.117],
+        [null, null, null],
+        [63.618, 6.359, 6.805],
+        [60.321, 2.953, 2.964],
+        [null, null, null],
+        [61.333, 4.652, 4.989],
+        [70.968, 13.933, 13.839],
+        [215.241, 156.281, 156.074],
+        [213.746, 167.137, 158.966],
+        [69.341, 12.259, 12.45],
+        [69.271, 13.099, 13.182],
+        [77.447, 20.517, 20.89],
+        [59.354, 1.714, 1.992],
+        [58.747, 1.413, 1.503],
+        [58.48, 0.919, 0.898],
+        [61.248, 3.528, 3.925],
+        [58.614, 0.99, 1.076],
+        [58.452, 0.903, 1.016],
+        [58.543, 0.914, 0.958]
+]
+  } 
+
diff --git a/dask/results/20261012/c7a.metal-48xl.json b/dask/results/20261012/c7a.metal-48xl.json
@@ -0,0 +1,60 @@
+{
+      "system": "Dask (DataFrame)",
+      "date": "2026-10-12",
+      "machine": "c7a.metal-48xl",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","column-oriented"],
+      "load_time": 5,
+      "data_size": 174168110180,
+      "concurrent_qps": 0.078,
+      "concurrent_error_ratio": 0.145,
+      "result": [
+        [63.748, 6.45, 6.535],
+        [58.011, 0.663, 0.687],
+        [57.669, 0.254, 0.193],
+        [57.511, 0.228, 0.143],
+        [58.447, 0.863, 0.905],
+        [59.197, 1.606, 1.84],
+        [57.228, 0.241, 0.138],
+        [58.061, 0.475, 0.53],
+        [59.365, 1.948, 2.04],
+        [null, null, null],
+        [66.479, 9.341, 9.178],
+        [67.631, 10.225, 10.47],
+        [61.172, 3.536, 3.811],
+        [66.428, 9.875, 9.686],
+        [111.985, 43.529, 42.93],
+        [58.671, 1.236, 1.393],
+        [135.47, 81.894, 85.483],
+        [63.181, 6.578, 6.654],
+        [139.566, 77.851, 86.547],
+        [57.671, 0.396, 0.405],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [58.327, 0.9, 0.784],
+        [null, null, null],
+        [62.09, 4.983, 5.216],
+        [60.008, 2.378, 2.468],
+        [null, null, null],
+        [60.277, 3.589, 4.01],
+        [72.089, 14.691, 14.88],
+        [233.177, 177.627, 178.588],
+        [235.036, 181.19, 182.43],
+        [68.465, 10.774, 11.097],
+        [68.138, 10.852, 11.317],
+        [77.554, 20.556, 20.747],
+        [58.586, 1.475, 1.323],
+        [58.575, 1.028, 1.295],
+        [57.935, 0.617, 0.741],
+        [60.562, 3.214, 3.351],
+        [58.375, 0.827, 0.901],
+        [58.112, 0.644, 0.779],
+        [58.158, 0.773, 0.732]
+]
+  } 
+
diff --git a/dask/results/20261012/c8g.metal-48xl.json b/dask/results/20261012/c8g.metal-48xl.json
@@ -0,0 +1,60 @@
+{
+      "system": "Dask (DataFrame)",
+      "date": "2026-10-12",
+      "machine": "c8g.metal-48xl",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","column-oriented"],
+      "load_time": 5,
+      "data_size": 174168110180,
+      "concurrent_qps": 0.105,
+      "concurrent_error_ratio": 0.16,
+      "result": [
+        [61.648, 4.543, 4.55],
+        [57.447, 0.546, 0.592],
+        [57.007, 0.221, 0.159],
+        [57.137, 0.174, 0.12],
+        [57.806, 0.819, 0.838],
+        [58.425, 1.328, 1.318],
+        [56.946, 0.136, 0.113],
+        [57.398, 0.378, 0.413],
+        [58.651, 1.692, 1.66],
+        [null, null, null],
+        [63.541, 6.516, 6.779],
+        [64.407, 7.567, 7.532],
+        [59.767, 3.064, 3.18],
+        [63.328, 6.618, 6.601],
+        [94.011, 37.187, 36.636],
+        [57.982, 1.101, 1.154],
+        [115.844, 58.468, 58.756],
+        [61.746, 4.767, 5.014],
+        [115.38, 64.369, 70.918],
+        [57.238, 0.347, 0.352],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [null, null, null],
+        [57.555, 0.633, 0.579],
+        [null, null, null],
+        [61.599, 4.962, 4.649],
+        [58.824, 1.921, 1.938],
+        [null, null, null],
+        [59.613, 2.875, 3.079],
+        [66.683, 9.846, 9.943],
+        [164.543, 106.876, 106.357],
+        [163.062, 106.104, 106.756],
+        [66.589, 9.604, 9.237],
+        [65.916, 10.05, 9.526],
+        [70.284, 13.38, 13.278],
+        [58.155, 1.176, 1.215],
+        [57.748, 0.829, 0.87],
+        [57.578, 0.59, 0.617],
+        [59.509, 2.641, 2.672],
+        [57.693, 0.652, 0.66],
+        [57.606, 0.607, 0.599],
+        [57.573, 0.673, 0.631]
+]
+  } 
+
diff --git a/pandas/queries.sql b/pandas/queries.sql
@@ -20,7 +20,7 @@ hits.groupby([hits['UserID'], hits['EventTime'].dt.minute, 'SearchPhrase']).size
 hits[hits['UserID'] == 435090932899640449]
 hits[hits['URL'].str.contains('google')].shape[0]
 hits[hits['URL'].str.contains('google') & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg({'URL': 'min', 'SearchPhrase': 'size'}).nlargest(10, 'SearchPhrase')
-hits[hits['Title'].str.contains('Google') & ~hits['URL'].str.contains('.google.') & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg({'URL': 'min', 'Title': 'min', 'SearchPhrase': 'size', 'UserID': 'nunique'}).nlargest(10, 'SearchPhrase')
+hits[hits['Title'].str.contains('Google') & ~hits['URL'].str.contains('.google.', regex=False) & (hits['SearchPhrase'] != '')].groupby('SearchPhrase').agg({'URL': 'min', 'Title': 'min', 'SearchPhrase': 'size', 'UserID': 'nunique'}).nlargest(10, 'SearchPhrase')
 hits[hits['URL'].str.contains('google')].sort_values(by='EventTime').head(10)
 hits[hits['SearchPhrase'] != ''].sort_values(by='EventTime')[['SearchPhrase']].head(10)
 hits[hits['SearchPhrase'] != ''].sort_values(by='SearchPhrase')[['SearchPhrase']].head(10)

diff --git a/pandas/results/20261012/c6a.metal.json b/pandas/results/20261012/c6a.metal.json
@@ -0,0 +1,60 @@
+{
+      "system": "pandas",
+      "date": "2026-10-12",
+      "machine": "c6a.metal",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","row-oriented"],
+      "load_time": 18,
+      "data_size": 81770409884,
+      "concurrent_qps": 0.017,
+      "concurrent_error_ratio": 0,
+      "result": [
+        [74.186, 4.431, 4.37],
+        [76.275, 7.783, 6.587],
+        [69.532, 0.064, 0.063],
+        [69.242, 0.058, 0.084],
+        [73.548, 3.839, 3.786],
+        [72.928, 2.366, 2.181],
+        [69.782, 0.321, 0.283],
+        [75.701, 6.282, 7.849],
+        [78.99, 9.504, 9.33],
+        [79.644, 10.316, 10.308],
+        [80.934, 13.67, 11.376],
+        [80.881, 11.749, 11.257],
+        [88.782, 21.954, 20.451],
+        [92.201, 24.737, 23.611],
+        [91.294, 23.469, 22.756],
+        [76.864, 7.438, 7.398],
+        [92.635, 22.369, 22.865],
+        [92.414, 22.687, 22.2],
+        [274.62, 207.574, 191.156],
+        [75.236, 6.917, 5.919],
+        [82.633, 13.437, 12.932],
+        [82.897, 13.802, 13.334],
+        [101.155, 34.86, 30.96],
+        [83.008, 13.579, 12.93],
+        [95.351, 28.796, 34.519],
+        [108.112, 42.981, 50.119],
+        [105.379, 40.05, 47.197],
+        [237.399, 163.698, 157.938],
+        [708.116, 617.828, 859.843],
+        [105.649, 36.12, 36.312],
+        [87.459, 19.133, 18.957],
+        [94.396, 25.894, 27.79],
+        [160.68, 107.285, 97.536],
+        [96.099, 26.847, 30.524],
+        [95.626, 26.691, 26.55],
+        [105.682, 37.625, 37.056],
+        [76.768, 7.469, 7.44],
+        [77.535, 6.853, 8.079],
+        [76.427, 7.686, 7.742],
+        [76.959, 8.015, 8.144],
+        [76.611, 8.105, 6.75],
+        [76.249, 7.419, 6.625],
+        [77.318, 8.743, 7.849]
+]
+  } 
+
diff --git a/pandas/results/20261012/c7a.metal-48xl.json b/pandas/results/20261012/c7a.metal-48xl.json
@@ -0,0 +1,60 @@
+{
+      "system": "pandas",
+      "date": "2026-10-12",
+      "machine": "c7a.metal-48xl",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","row-oriented"],
+      "load_time": 15,
+      "data_size": 81770409884,
+      "concurrent_qps": 0.023,
+      "concurrent_error_ratio": 0,
+      "result": [
+        [71.718, 4.28, 4.276],
+        [72.484, 6.055, 5.989],
+        [66.936, 0.065, 0.058],
+        [66.979, 0.057, 0.058],
+        [71.079, 4.329, 4.035],
+        [69.641, 2.345, 2.167],
+        [68.02, 0.215, 0.17],
+        [72.025, 6.092, 4.944],
+        [76.019, 9.344, 9.183],
+        [77.18, 10.238, 10.358],
+        [76.377, 9.47, 9.333],
+        [76.97, 9.798, 9.459],
+        [84.988, 20.113, 18.507],
+        [87.652, 21.768, 20.906],
+        [87.493, 21.812, 20.451],
+        [76.247, 9.527, 9.564],
+        [92.186, 25.419, 25.159],
+        [92.396, 25.435, 25.666],
+        [254.356, 186.598, 186.666],
+        [71.857, 5.389, 5.288],
+        [77.979, 10.982, 10.663],
+        [78.342, 11.145, 11.15],
+        [95.159, 28.504, 29.406],
+        [77.996, 11.011, 10.751],
+        [90.275, 24.827, 23.521],
+        [102.555, 37.984, 38.862],
+        [100.619, 36.713, 35.635],
+        [299.99, 304.605, 239.398],
+        [610.325, 561.153, 529.898],
+        [92.114, 24.068, 24.043],
+        [83.525, 16.753, 16.581],
+        [92.008, 26.075, 25.238],
+        [187.692, 131.186, 123.959],
+        [92.875, 25.927, 25.39],
+        [93.01, 26.097, 25.389],
+        [108.506, 41.951, 42.308],
+        [72.943, 5.822, 5.832],
+        [72.618, 6.353, 5.289],
+        [72.551, 5.803, 4.781],
+        [72.78, 5.777, 5.679],
+        [72.525, 6.119, 5.089],
+        [72.094, 5.903, 5.946],
+        [73.416, 6.961, 6.938]
+]
+  } 
+
diff --git a/pandas/results/20261012/c8g.metal-48xl.json b/pandas/results/20261012/c8g.metal-48xl.json
@@ -0,0 +1,60 @@
+{
+      "system": "pandas",
+      "date": "2026-10-12",
+      "machine": "c8g.metal-48xl",
+      "cluster_size": 1,
+      "proprietary": "no",
+      "hardware": "cpu",
+      "tuned": "no",
+      "tags": ["Python","dataframe","in-memory","row-oriented"],
+      "load_time": 12,
+      "data_size": 81770409884,
+      "concurrent_qps": 0.03,
+      "concurrent_error_ratio": 0,
+      "result": [
+        [72.916, 8.277, 8.285],
+        [67.882, 3.467, 3.461],
+        [64.329, 0.142, 0.14],
+        [64.106, 0.053, 0.053],
+        [67.662, 2.59, 2.602],
+        [67.318, 2.066, 2.066],
+        [64.514, 0.248, 0.248],
+        [67.988, 3.487, 3.473],
+        [70.721, 6.215, 6.218],
+        [71.223, 7.02, 7.027],
+        [73.199, 8.21, 8.104],
+        [72.418, 7.962, 7.906],
+        [80.482, 16.043, 16.014],
+        [81.167, 16.758, 16.831],
+        [81.41, 17.107, 17.145],
+        [70.938, 6.383, 6.379],
+        [84.449, 19.395, 19.433],
+        [83.748, 19.312, 19.28],
+        [228.24, 163.038, 163.529],
+        [67.729, 2.844, 2.842],
+        [75.407, 10.97, 11.006],
+        [75.8, 11.316, 11.319],
+        [94.403, 29.94, 29.601],
+        [76.048, 11.017, 10.997],
+        [87.329, 22.901, 22.983],
+        [98.614, 34.465, 34.139],
+        [97.218, 32.058, 32.132],
+        [190.194, 124.234, 155.564],
+        [577.368, 513.725, 511.243],
+        [84.965, 20.327, 20.119],
+        [77.17, 12.091, 12.054],
+        [83.009, 18.651, 18.677],
+        [140.268, 77.016, 78.146],
+        [90.178, 25.387, 25.401],
+        [90.229, 25.362, 25.386],
+        [93.988, 29.823, 29.911],
+        [68.371, 3.761, 3.767],
+        [69.743, 4.842, 4.799],
+        [69.8, 4.531, 4.508],
+        [68.311, 3.879, 3.877],
+        [68.263, 3.765, 3.745],
+        [69.672, 4.962, 4.924],
+        [69.771, 5.304, 5.301]
+]
+  } 
+
diff --git a/polars-dataframe/queries.sql b/polars-dataframe/queries.sql
@@ -20,7 +20,7 @@ hits.group_by([pl.col('UserID'), pl.col('EventTime').dt.minute(), 'SearchPhrase'
 hits.select('UserID').filter(pl.col('UserID') == 435090932899640449).collect()
 hits.filter(pl.col('URL').str.contains('google')).select(pl.len()).collect().item()
 hits.filter(pl.col('URL').str.contains('google') & (pl.col('SearchPhrase') != '')).group_by('SearchPhrase').agg([pl.col('URL').min(), pl.len().alias('count')]).sort('count', descending=True).head(10).collect()
-hits.filter(pl.col('Title').str.contains('Google') & ~pl.col('URL').str.contains('.google.') & (pl.col('SearchPhrase') != '')).group_by('SearchPhrase').agg([pl.col('URL').min(), pl.col('Title').min(), pl.len().alias('count'), pl.col('UserID').n_unique()]).sort('count', descending=True).head(10).collect()
+hits.filter(pl.col('Title').str.contains('Google') & ~pl.col('URL').str.contains('.google.', literal=True) & (pl.col('SearchPhrase') != '')).group_by('SearchPhrase').agg([pl.col('URL').min(), pl.col('Title').min(), pl.len().alias('count'), pl.col('UserID').n_unique()]).sort('count', descending=True).head(10).collect()
 hits.filter(pl.col('URL').str.contains('google')).sort('EventTime').head(10).collect()
 hits.filter(pl.col('SearchPhrase') != '').sort('EventTime').select('SearchPhrase').head(10).collect()
 hits.filter(pl.col('SearchPhrase') != '').sort('SearchPhrase').select('SearchPhrase').head(10).collect()