Add Exasol benchmark support with TPC-H queries

Makefile

@@ -10,6 +10,9 @@ NUM_PARTITIONS=10
 NUM_BATCHES?=1  ## data split into this number of batches, more batches reduce disk space required for temporary tbl files
 PARALLELISM?=8  ## number of parallel data generation processes, can be 1, unless NUM_BATCHES is 1
 
+# Default scale factor for data generation (set to override via environment)
+SCALE_FACTOR ?= 1.0
+
 .venv:  ## Set up Python virtual environment and install dependencies
 	python3 -m venv $(VENV)
 	$(MAKE) install-deps
@@ -63,7 +66,6 @@ data/tables/scale-$(SCALE_FACTOR): .venv  ## Generate data tables
 	# mkdir -p "data/tables/scale-$(SCALE_FACTOR)"
 	# mv tpch-dbgen/*.tbl data/tables/scale-$(SCALE_FACTOR)/
 	# $(VENV_BIN)/python -m scripts.prepare_data --tpch_gen_folder="data/tables/scale-$(SCALE_FACTOR)"
-	rm -rf data/tables/scale-$(SCALE_FACTOR)/*.tbl
 
 .PHONY: data-tables-partitioned
 data-tables-partitioned: data/tables/scale-$(SCALE_FACTOR)/${NUM_PARTITIONS}
@@ -113,8 +115,12 @@ run-dask: .venv data-tables ## Run Dask benchmarks
 run-modin: .venv data-tables ## Run Modin benchmarks
 	$(VENV_BIN)/python -m queries.modin
 
+.PHONY: run-exasol
+run-exasol: .venv data-tables ## Run Exasol benchmarks
+	$(VENV_BIN)/python -m queries.exasol
+
 .PHONY: run-all
-run-all: run-polars run-duckdb run-pandas run-pyspark run-dask run-modin  ## Run all benchmarks
+run-all: run-polars run-duckdb run-pandas run-pyspark run-dask run-modin run-exasol  ## Run all benchmarks
 
 .PHONY: plot
 plot: .venv  ## Plot results

README.md

@@ -43,3 +43,30 @@ This will do the following,
 - Create a new virtual environment with all required dependencies.
 - Generate data for benchmarks.
 - Run the benchmark suite.
+
+> **Note:** To run the Exasol benchmarks, ensure your Exasol database is configured via environment variables (or a `.env` file) with the following settings:
+>
+> ```
+> EXASOL_HOST=<host>
+> EXASOL_USER=<user>
+> EXASOL_PASSWORD=<password>
+> (optional) EXASOL_PORT=<port>          # default: 8563
+> (optional) EXASOL_SCHEMA_NAME=<schema> # default: tpc
+> ```
+>
+> You can set SCALE_FACTOR=<scale> (default: 1.0) to control the data scale.
+> Then execute:
+>
+> ```shell
+> make run-exasol
+> ```
+>
+> The Exasol runner executes the DDL scripts in three phases:
+>
+> 1. `create_schema.sql` to create empty TPC-H tables.
+> 2. Import data files into these tables from the local filesystem using Exasol's
+>    `IMPORT FROM` functionality (via `pyexasol.import_from_file`).
+> 3. Run `create_indices_1node.sql` and `analyze_database.sql` to enforce indices
+>    and collect statistics.
+>
+> Finally, it executes the 22 TPC-H queries.

queries/exasol/__main__.py

@@ -0,0 +1,12 @@
+from queries.common_utils import execute_all
+from queries.exasol.utils import prepare_load_data, prepare_postload, prepare_schema
+
+if __name__ == "__main__":
+    # create schema before loading data
+    prepare_schema()
+    # load data files into Exasol tables
+    prepare_load_data()
+    # enforce indices and collect statistics after data load
+    prepare_postload()
+    # run the TPC-H queries
+    execute_all("exasol")

queries/exasol/q1.py

@@ -0,0 +1,38 @@
+from queries.exasol import utils
+
+Q_NUM = 1
+
+
+def q() -> None:
+
+    line_item_ds = utils.get_line_item_ds()
+    query_str = f"""
+    select
+            l_returnflag,
+            l_linestatus,
+            sum(l_quantity) as sum_qty,
+            sum(l_extendedprice) as sum_base_price,
+            sum(l_extendedprice * (1 - l_discount)) as sum_disc_price,
+            sum(l_extendedprice * (1 - l_discount) * (1 + l_tax)) as sum_charge,
+            avg(l_quantity) as avg_qty,
+            avg(l_extendedprice) as avg_price,
+            avg(l_discount) as avg_disc,
+            count(*) as count_order
+        from
+            {line_item_ds}
+        where
+            l_shipdate <= '1998-09-02'
+        group by
+            l_returnflag,
+            l_linestatus
+        order by
+            l_returnflag,
+            l_linestatus
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q10.py

@@ -0,0 +1,52 @@
+from queries.exasol import utils
+
+Q_NUM = 10
+
+
+def q() -> None:
+
+    customer_ds = utils.get_customer_ds()
+    line_item_ds = utils.get_line_item_ds()
+    nation_ds = utils.get_nation_ds()
+    orders_ds = utils.get_orders_ds()
+    query_str = f"""
+    select
+            c_custkey,
+            c_name,
+            round(sum(l_extendedprice * (1 - l_discount)), 2) as revenue,
+            c_acctbal,
+            n_name,
+            c_address,
+            c_phone,
+            c_comment
+        from
+            {customer_ds},
+            {orders_ds},
+            {line_item_ds},
+            {nation_ds}
+        where
+            c_custkey = o_custkey
+            and l_orderkey = o_orderkey
+            and o_orderdate >= date '1993-10-01'
+            and o_orderdate < date '1993-10-01' + interval '3' month
+            and l_returnflag = 'R'
+            and c_nationkey = n_nationkey
+        group by
+            c_custkey,
+            c_name,
+            c_acctbal,
+            c_phone,
+            n_name,
+            c_address,
+            c_comment
+        order by
+            revenue desc
+        limit 20
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q11.py

@@ -0,0 +1,46 @@
+from queries.exasol import utils
+
+Q_NUM = 11
+
+
+def q() -> None:
+
+    nation_ds = utils.get_nation_ds()
+    part_supp_ds = utils.get_part_supp_ds()
+    supplier_ds = utils.get_supplier_ds()
+    query_str = f"""
+    select
+            ps_partkey,
+            round(sum(ps_supplycost * ps_availqty), 2) as "value"
+        from
+            {part_supp_ds},
+            {supplier_ds},
+            {nation_ds}
+        where
+            ps_suppkey = s_suppkey
+            and s_nationkey = n_nationkey
+            and n_name = 'GERMANY'
+        group by
+            ps_partkey having
+                    sum(ps_supplycost * ps_availqty) > (
+                select
+                    sum(ps_supplycost * ps_availqty) * 0.0001
+                from
+                    {part_supp_ds},
+                    {supplier_ds},
+                    {nation_ds}
+                where
+                    ps_suppkey = s_suppkey
+                    and s_nationkey = n_nationkey
+                    and n_name = 'GERMANY'
+                )
+            order by
+                "value" desc
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q12.py

@@ -0,0 +1,46 @@
+from queries.exasol import utils
+
+Q_NUM = 12
+
+
+def q() -> None:
+
+    line_item_ds = utils.get_line_item_ds()
+    orders_ds = utils.get_orders_ds()
+    query_str = f"""
+    select
+            l_shipmode,
+            sum(case
+                when o_orderpriority = '1-URGENT'
+                    or o_orderpriority = '2-HIGH'
+                    then 1
+                else 0
+            end) as high_line_count,
+            sum(case
+                when o_orderpriority <> '1-URGENT'
+                    and o_orderpriority <> '2-HIGH'
+                    then 1
+                else 0
+            end) as low_line_count
+        from
+            {orders_ds},
+            {line_item_ds}
+        where
+            o_orderkey = l_orderkey
+            and l_shipmode in ('MAIL', 'SHIP')
+            and l_commitdate < l_receiptdate
+            and l_shipdate < l_commitdate
+            and l_receiptdate >= date '1994-01-01'
+            and l_receiptdate < date '1994-01-01' + interval '1' year
+        group by
+            l_shipmode
+        order by
+            l_shipmode
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q13.py

@@ -0,0 +1,36 @@
+from queries.exasol import utils
+
+Q_NUM = 13
+
+
+def q() -> None:
+
+    customer_ds = utils.get_customer_ds()
+    orders_ds = utils.get_orders_ds()
+    query_str = f"""
+    select
+            c_count, count(*) as custdist
+        from (
+        select
+            c_custkey,
+            count(o_orderkey) as c_count
+            from
+                {customer_ds} left outer join {orders_ds} on
+                c_custkey = o_custkey
+                and o_comment not like '%special%requests%'
+            group by
+                c_custkey
+            ) c_orders
+        group by
+            c_count
+        order by
+            custdist desc,
+            c_count desc
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q14.py

@@ -0,0 +1,31 @@
+from queries.exasol import utils
+
+Q_NUM = 14
+
+
+def q() -> None:
+
+    line_item_ds = utils.get_line_item_ds()
+    part_ds = utils.get_part_ds()
+    query_str = f"""
+    select
+            round(100.00 * sum(case
+                when p_type like 'PROMO%'
+                    then l_extendedprice * (1 - l_discount)
+                else 0
+            end) / sum(l_extendedprice * (1 - l_discount)), 2) as promo_revenue
+        from
+            {line_item_ds},
+            {part_ds}
+        where
+            l_partkey = p_partkey
+            and l_shipdate >= date '1995-09-01'
+            and l_shipdate < date '1995-09-01' + interval '1' month
+        ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+

queries/exasol/q15.py

@@ -0,0 +1,39 @@
+from queries.exasol import utils
+
+Q_NUM = 15
+
+
+def q() -> None:
+
+    supplier_ds = utils.get_supplier_ds()
+    line_item_ds = utils.get_line_item_ds()
+    query_str = f"""
+    WITH revenue AS (
+        SELECT
+            l_suppkey AS supplier_no,
+            SUM(l_extendedprice * (1 - l_discount)) AS total_revenue
+        FROM {line_item_ds}
+        WHERE l_shipdate >= DATE '1996-01-01'
+          AND l_shipdate < DATE '1996-01-01' + INTERVAL '3' MONTH
+        GROUP BY l_suppkey
+    )
+    SELECT
+        s_suppkey,
+        s_name,
+        s_address,
+        s_phone,
+        total_revenue
+    FROM {supplier_ds}, revenue
+    WHERE s_suppkey = supplier_no
+      AND total_revenue = (
+          SELECT MAX(total_revenue) FROM revenue
+      )
+    ORDER BY s_suppkey
+    ;
+    """
+    utils.run_query(Q_NUM, query_str)
+
+
+if __name__ == "__main__":
+    q()
+