Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

DataFusion v33 update and add benchmark for partitioned parquet #145

Merged
merged 8 commits into from
Dec 2, 2023
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
90 changes: 45 additions & 45 deletions datafusion/results/single.json
Original file line number Diff line number Diff line change
@@ -1,58 +1,58 @@
{
"system": "DataFusion (Parquet, single)",
"date": "2023-07-29",
"date": "2023-11-19",
"machine": "c6a.4xlarge, 500gb gp2",
"cluster_size": 1,
"comment": "v28.0.0 (51b4392)",
"comment": "v33.0.0 (d2efaa9)",

"tags": ["Rust", "column-oriented", "embedded", "stateless"],

"load_time": 0,
"data_size": 14779976446,

"result": [
[2.641, 0.232, 0.216],
[0.092, 0.074, 0.073],
[0.159, 0.121, 0.117],
[0.351, 0.122, 0.120],
[1.142, 0.874, 0.869],
[1.399, 1.303, 1.329],
[0.096, 0.075, 0.076],
[0.094, 0.074, 0.076],
[1.516, 1.481, 1.475],
[2.711, 2.582, 2.583],
[0.429, 0.332, 0.327],
[0.558, 0.357, 0.360],
[1.374, 1.323, 1.336],
[3.559, 2.670, 2.733],
[1.519, 1.487, 1.477],
[1.041, 0.981, 0.988],
[3.248, 2.748, 2.801],
[3.139, 2.696, 2.688],
[7.014, 5.770, 5.775],
[0.271, 0.111, 0.109],
[9.975, 1.558, 1.578],
[11.163, 1.926, 1.881],
[22.053, 4.208, 4.194],
[56.007, 12.132, 12.113],
[2.559, 0.602, 0.586],
[0.737, 0.479, 0.489],
[2.548, 0.589, 0.589],
[9.550, 2.299, 2.286],
[9.076, 5.399, 5.388],
[0.579, 0.568, 0.584],
[2.214, 1.136, 1.136],
[5.734, 1.593, 1.598],
[8.357, 7.896, 8.053],
[11.556, 7.358, 7.369],
[12.051, 7.878, 7.956],
[1.866, 1.820, 1.807],
[0.448, 0.347, 0.358],
[0.231, 0.197, 0.195],
[0.248, 0.198, 0.188],
[0.843, 0.725, 0.722],
[0.138, 0.091, 0.095],
[0.115, 0.089, 0.088],
[0.124, 0.097, 0.092]
[0.076, 0.045, 0.045],
[0.110, 0.061, 0.062],
[0.200, 0.109, 0.106],
[1.529, 0.117, 0.113],
[2.265, 0.832, 0.836],
[2.652, 1.246, 1.247],
[0.091, 0.065, 0.071],
[0.100, 0.063, 0.063],
[2.465, 1.490, 1.478],
[3.672, 2.539, 2.496],
[1.627, 0.344, 0.330],
[1.885, 0.366, 0.369],
[2.679, 1.253, 1.236],
Copy link
Contributor

@Dandandan Dandandan Nov 29, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Quite some queries regressed, are you sure you are using the same setup as @jychen7
in #127
(c6a.4xlarge, 500gb gp2) ?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it was gp3 storage actually, which is what new instances default to. let me re-run this again and confirm the gp2 500gb storage. I'll put this in draft until I update this and for nickitat's comment.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

definitely was gp3 previously. rebuilt with gp2 500gb, results more in line with expectations.

[6.053, 2.504, 2.505],
[2.885, 1.386, 1.382],
[1.776, 0.968, 0.927],
[5.752, 2.709, 2.689],
[5.639, 2.646, 2.629],
[10.707, 5.692, 5.799],
[1.171, 0.100, 0.105],
[20.537, 1.570, 1.575],
[23.093, 1.901, 1.897],
[44.658, 4.103, 4.065],
[112.447, 11.674, 11.654],
[5.791, 0.581, 0.558],
[2.340, 0.511, 0.529],
[5.823, 0.626, 0.632],
[19.944, 2.333, 2.343],
[17.608, 5.052, 5.086],
[0.557, 0.500, 0.494],
[4.888, 1.140, 1.126],
[11.791, 1.553, 1.566],
[12.415, 7.764, 7.724],
[21.279, 6.722, 6.816],
[21.569, 7.325, 7.297],
[1.981, 1.839, 1.821],
[0.466, 0.306, 0.312],
[0.254, 0.185, 0.182],
[0.284, 0.173, 0.171],
[0.887, 0.685, 0.667],
[0.171, 0.092, 0.095],
[0.144, 0.086, 0.087],
[0.151, 0.101, 0.097]
]
}
Loading