The TPCH database and dbgen data generation utility, courtesy of www.tpc.org, were developed to provide an approach to benchmarking and include:

• The tpch Database structure
• The tpch dbgen utility, a utility to populate the database with a specified amount of data (Scale Factor)
• The tpch benchmark queries, a set of pre-defined data warehouse queries to run against the database

We will show the details of the creation of the tpch database and it's population using the dbgen utility to generate data.

In essence, the schema consists of 8 tables, 8 explicit unique indexes supporting 8 primary keys and 9 explicit indexes supporting 9 foreign keys.

The tpch dbgen utility generates, by default, a set of flat files suitable for loading into the tpch schema with the size based on the “Scale Factor” argument. A scale factor of 1 produces a complete data set of approximately 1 GB, a scale factor of 10 produces a data set of approximately 10 GB etc.

Download the dbgen source code:

$ git clone https://github.com/electrum/tpch-dbgen.git

Cloning into 'tpch-dbgen'...
remote: Counting objects: 149, done.
remote: Total 149 (delta 0), reused 0 (delta 0), pack-reused 149
Receiving objects: 100% (149/149), 216.15 KiB | 202.00 KiB/s, done.
Resolving deltas: 100% (30/30), done.
Checking connectivity... done.

In the downloaded directory (tpch-dbgen), edit the file makefile.suite and set the following variables to the appropriate vaules:

CC=gcc
DATABASE=INFORMIX
MACHINE=LINUX
WORKLOAD=TPCH

The run the make utility:

$ make -f makefile.suite

gcc -g -DDBNAME=\"dss\" -DMAC -DINFORMIX -DTPCH -DRNG_TEST -D_FILE_OFFSET_BITS=64    -c -o build.o build.c
gcc -g -DDBNAME=\"dss\" -DMAC -DINFORMIX -DTPCH -DRNG_TEST -D_FILE_OFFSET_BITS=64    -c -o driver.o driver.c
...

Now you are ready to generate the tpch files

• Change to the appropriate directory where you want to generate tpch files. For example, create a subdirectory under the tpch-dbgen directory.
  Copy

  $ mkdir data
  $ cd data

• Copy the dbgen executable file and dists.dss file there.
  Copy

  $ cp ../dbgen .
  $ cp ../dists.dss .

• Run dbgen for the appropriate database size factor (1GB in the sample).
  Copy

  ./dbgen -s 1

• Generation may take a while. When completed, you can see the resulting files.
  Copy

  $ ls -l

  total 2150000
  -rw-r--r--  1 deister  staff   24346144 13 may 12:05 customer.tbl
  -rw-r--r--  1 deister  staff  759863287 13 may 12:05 lineitem.tbl
  -rw-r--r--  1 deister  staff       2224 13 may 12:05 nation.tbl
  -rw-r--r--  1 deister  staff  171952161 13 may 12:05 orders.tbl
  -rw-r--r--  1 deister  staff   24135125 13 may 12:05 part.tbl
  -rw-r--r--  1 deister  staff  118984616 13 may 12:05 partsupp.tbl
  -rw-r--r--  1 deister  staff        389 13 may 12:05 region.tbl
  -rw-r--r--  1 deister  staff    1409184 13 may 12:05 supplier.tbl

Database will be sized according the selected scale factor.