8000 chore: round earlier in TPC-H q15 to try and reduce non-determinism due to aggregating twice by tswast · Pull Request #1877 · googleapis/python-bigquery-dataframes · GitHub
[go: up one dir, main page]
Skip to content

chore: round earlier in TPC-H q15 to try and reduce non-determinism due to aggregating twice #1877

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.

Sign up for GitHub

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

Merged
tswast merged 3 commits into from
Jul 7, 2025
Merged

chore: round earlier in TPC-H q15 to try and reduce non-determinism due to aggregating twice #1877

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d4c5b77
chore: round earlier in TPC-H q15 to try and reduce non-determinism d…
tswast Jul 1, 2025
ebb11d8
remove unnecessary code
tswast Jul 7, 2025
2c31693
Merge branch 'main' into b384588336-tpc-h-q15-non-determinism
tswast Jul 7, 2025
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
9 changes: 5 additions & 4 deletions third_party/bigframes_vendored/tpch/queries/q15.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,11 @@ def q(project_id: str, dataset_id: str, session: bigframes.Session):
.agg(TOTAL_REVENUE=bpd.NamedAgg(column="REVENUE", aggfunc="sum"))
.rename(columns={"L_SUPPKEY": "SUPPLIER_NO"})
)
# Round earlier to prevent non-determinism in the later join due to
# differences in distributed floating point operation sort order.
grouped_revenue = grouped_revenue.assign(
TOTAL_REVENUE=grouped_revenue["TOTAL_REVENUE"].round(2)
)

joined_data = bpd.merge(
supplier, grouped_revenue, left_on="S_SUPPKEY", right_on="SUPPLIER_NO"
Expand All @@ -43,10 +48,6 @@ def q(project_id: str, dataset_id: str, session: bigframes.Session):
max_revenue_suppliers = joined_data[
joined_data["TOTAL_REVENUE"] == joined_data["MAX_REVENUE"]
]

max_revenue_suppliers["TOTAL_REVENUE"] = max_revenue_suppliers[
"TOTAL_REVENUE"
].round(2)
q_final = max_revenue_suppliers[
["S_SUPPKEY", "S_NAME", "S_ADDRESS", "S_PHONE", "TOTAL_REVENUE"]
].sort_values("S_SUPPKEY")
Expand Down
0