Run TPCH with one command

Install the Trino TPCH generator on every FE and BE using the shared installation instructions, then configure conf/doris-cluster.conf and run:

./bin/run-tpch.sh -s 1 -d tpch_sf1

This streams generated rows directly into Doris, creates the existing benchmark tables, collects statistics, and records three executions per query. It supports SF1, SF100, SF1000, and SF10000. Preparation requires a new database. Use --queries-only to measure existing data again, with the same scale factor.

See shared workflow options, result files, and timing semantics. TPCH uses the original scale-specific table DDL and multi-table queries. Install the plugin before running the command.

Original file-based workflow

Usage

These scripts are used to make tpc-h test. follow the steps below:

1. build tpc-h dbgen tool.

./bin/build-tpch-dbgen.sh

If the build failed in dbgen tools' compilation, update your GCC version or change download link used by wget in build-tpch-dbgen.sh from ".../TPC-H_Tools_v3.0.0new.zip" to ".../TPC-H_Tools_v3.0.0.zip"

2. generate tpc-h data. use -h for more infomations.

./bin/gen-tpch-data.sh -s 1

3. create tpc-h tables. modify conf/doris-cluster.conf to specify doris info, then run script below.

./bin/create-tpch-tables.sh -s 1

4. load tpc-h data. use -h for help.

./bin/load-tpch-data.sh

5. run tpc-h queries.

./bin/run-tpch-queries.sh -s 1

NOTICE: At present, Doris's query optimizer and statistical information functions are not complete, so we rewrite some queries in TPC-H to adapt to Doris' execution framework, but it does not affect the correctness of the results. The rewritten SQL is marked with "Modified" in the corresponding .sql file.

A new query optimizer will be released in subsequent releases.

Currently, differnt scales use the same suite of query sqls.