Chapter 4. System requirements

Procedure

1. Check if the extension is available on the PostgreSQL server (automation hub database).

   $ psql -d <automation hub database> -c "SELECT * FROM pg_available_extensions WHERE name='hstore'"

   Where the default value for <automation hub database> is automationhub.

   This gives an output similar to the following:

   name  | default_version | installed_version |comment
   ------+-----------------+-------------------+---------------------------------------------------
    hstore | 1.7           |                   | data type for storing sets of (key, value) pairs
   (1 row)

2. This indicates that the hstore 1.7 extension is available, but not enabled.

   If the hstore extension is not available on the PostgreSQL server, the result is similar to the following:

    name | default_version | installed_version | comment
   ------+-----------------+-------------------+---------
   (0 rows)

3. On a RHEL based server, the hstore extension is included in the postgresql-contrib RPM package, which is not installed automatically when installing the PostgreSQL server RPM package.

   To install the RPM package, use the following command:

   dnf install postgresql-contrib

4. Create the hstore PostgreSQL extension on the automation hub database with the following command:

   $ psql -d <automation hub database> -c "CREATE EXTENSION hstore"

   The output of which is:

   CREATE EXTENSION

5. In the following output, the installed_version field contains the hstore extension used, indicating that hstore is enabled.

   name | default_version | installed_version | comment
   -----+-----------------+-------------------+------------------------------------------------------
   hstore  |     1.7      |       1.7         | data type for storing sets of (key, value) pairs
   (1 row)

Procedure

1. Run a random write test:

   $ fio --name=write_iops --directory=/tmp --numjobs=3 --size=10G \
   --time_based --runtime=60s --ramp_time=2s --ioengine=libaio --direct=1 \
   --verify=0 --bs=4K --iodepth=64 --rw=randwrite \
   --group_reporting=1 > /tmp/fio_benchmark_write_iops.log \
   2>> /tmp/fio_write_iops_error.log

2. Run a random read test:

   $ fio --name=read_iops --directory=/tmp \
   --numjobs=3 --size=10G --time_based --runtime=60s --ramp_time=2s \
   --ioengine=libaio --direct=1 --verify=0 --bs=4K --iodepth=64 --rw=randread \
   --group_reporting=1 > /tmp/fio_benchmark_read_iops.log \
   2>> /tmp/fio_read_iops_error.log

3. Review the results:

   In the log files written by the benchmark commands, search for the line beginning with iops. This line shows the minimum, maximum, and average values for the test.

   The following example shows the line in the log file for the random read test:

   $ cat /tmp/fio_benchmark_read_iops.log
   read_iops: (g=0): rw=randread, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
   […]
      iops        : min=50879, max=61603, avg=56221.33, stdev=679.97, samples=360
   […]

   You must review, monitor, and revisit the log files according to your own business requirements, application workloads, and new demands.