TPC-DS Q90 — Web Sales AM/PM Ratio Analysis

TPC-DS benchmark · all TPC-DS queries

Computes the ratio of AM to PM web sales counts

Representative SQL

Rendered for DataFusion with default parameters.

SELECT
  CAST(amc AS DECIMAL(15, 4)) / NULLIF(CAST(pmc AS DECIMAL(15, 4)), 0) AS am_pm_ratio
FROM (
  SELECT
    COUNT(*) AS amc
  FROM web_sales, household_demographics, time_dim, web_page
  WHERE
    ws_sold_time_sk = time_dim.t_time_sk
    AND ws_ship_hdemo_sk = household_demographics.hd_demo_sk
    AND ws_web_page_sk = web_page.wp_web_page_sk
    AND time_dim.t_hour BETWEEN 6 AND 6 + 1
    AND household_demographics.hd_dep_count = 8
    AND web_page.wp_char_count BETWEEN 5000 AND 5200
) AS at, (
  SELECT
    COUNT(*) AS pmc
  FROM web_sales, household_demographics, time_dim, web_page
  WHERE
    ws_sold_time_sk = time_dim.t_time_sk
    AND ws_ship_hdemo_sk = household_demographics.hd_demo_sk
    AND ws_web_page_sk = web_page.wp_web_page_sk
    AND time_dim.t_hour BETWEEN 14 AND 14 + 1
    AND household_demographics.hd_dep_count = 8
    AND web_page.wp_char_count BETWEEN 5000 AND 5200
) AS pt
ORDER BY
  am_pm_ratio
LIMIT 100

Representative DataFrame

Expression family (Polars, PySpark, DataFusion), default parameters.

def q90_expression_impl(ctx: DataFrameContext) -> Any:
    hour_am, hour_pm, dep_count, char_min, char_max = _q90_params()
    col = ctx.col
    lit = ctx.lit
    base = (
        ctx.get_table("web_sales")
        .join(ctx.get_table("time_dim"), left_on="ws_sold_time_sk", right_on="t_time_sk")
        .join(ctx.get_table("household_demographics"), left_on="ws_ship_hdemo_sk", right_on="hd_demo_sk")
        .join(ctx.get_table("web_page"), left_on="ws_web_page_sk", right_on="wp_web_page_sk")
        .filter((col("hd_dep_count") == lit(dep_count)) & col("wp_char_count").is_between(char_min, char_max))
    )
    am_val = base.filter(col("t_hour").is_between(hour_am, hour_am + 1)).select(ctx.count().alias("amc")).scalar(0, 0)
    pm_val = base.filter(col("t_hour").is_between(hour_pm, hour_pm + 1)).select(ctx.count().alias("pmc")).scalar(0, 0)
    return ctx.create_dataframe({"am_pm_ratio": [am_val / pm_val if pm_val and pm_val > 0 else None]})

Get the exact query for your run

The blocks above are representative. To get the precise statement a specific platform runs at a specific scale, substitute your own values:

from benchbox import TPCDS

bench = TPCDS(scale_factor=1.0)            # your scale factor
queries = bench.get_queries(dialect="datafusion")   # your platform's dialect; drop dialect= for native SQL
print(queries['90'])

Parameter substitution for this benchmark is seed-driven; pass a seed for a specific stream’s parameters:

print(bench.get_query(90, scale_factor=1.0, seed=42, dialect="datafusion"))

For the DataFrame form:

from benchbox.core.query_catalog import get_dataframe_render

print(get_dataframe_render('tpcds', '90', family="expression").source)  # or family="pandas"

Source

Query text: _sources/tpc-ds/query_templates/query90.tpl