Developer Guide

  • 2021.2
  • 06/11/2021
  • Public
Contents

Step 4: Production: Apply Tuning Configuration

Imagine that you are now in the production phase. You will use the tuning configuration file generated in the preproduction phase to tune a production environment. For this demo, you will walk through the data streams optimizer production workflow, using the same target board as a proxy for a production environment.
  1. On the host system, go to the tools directory.
    cd ${TCC_TOOLS_PATH}
  2. Run the production tool to apply the tuning configuration to the target board. Replace
    <target_hostname>
    with the target system IP address or hostname and
    <date>
    as appropriate. Add the
    --username
    argument to specify the username (username is
    root
    by default). If the target board has a password, add the
    --password
    argument.
    python3 tcc_data_streams_optimizer_prod.py --apply <target_hostname>/single_corepcierd_1_<date>/tuning_configuration.json --hostname <target_hostname> --reboot --username <username> --password <password>
    where
    Option
    Description
    --apply <target_hostname>/single_corepcierd_1_<date>/tuning_configuration.json
    Apply the tuning from the specified configuration file.
    --hostname <target_hostname>
    Host name of the target system to be tuned (default: localhost).
    --reboot
    Reboot the target after applying the capsule. A reboot is required to apply the tuning configuration to the target system.
    --username <username>
    User name for the target system to be tuned (default: root).
    --password <password>
    Password for the target system to be tuned (empty by default).
  3. The tool applies the configuration. Confirm that you see output similar to the example below.
    Connected. Copying capsule(s) from host to target <target_hostname> Capsule(s) were copied Applying capsule(s) for <target_hostname>... Capsule(s) were applied for <target_hostname>. Configuration ready. Rebooting <target_hostname>...
  4. To verify that the configuration was applied successfully, use the same command as previous but change
    --apply
    to
    --verify
    , and remove
    --reboot
    .
    python3 ${TCC_ROOT}/tools/tcc_data_streams_optimizer_prod.py --verify <target_hostname>/single_corepcierd_1_<date>/tuning_configuration.json --hostname <target_hostname> --username <username> --password <password>
    where
    Option
    Description
    --verify /usr/share/tcc_tools/<target_hostname>/single_corepcierd_1_<date>/tuning_configuration.json
    Verify the current tuning against the specified configuration file.
    --hostname <target_hostname>
    Host name of the target system to be tuned (default: localhost).
  5. The tool compares the tuning configuration file to the target’s current configuration. Confirm that you see output similar to the example below.
    Connected. Verifying that configuration was applied successfully... Configuration applied.

Next Steps

You have confirmed that the tool is functioning properly. For your use case and workload, you will need to create your own input files, such as a workload validation script, requirements file, and environment file. Learn more about customizing the input files for your workload, starting in Create Input Files for Your Workload.

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.