Developer Guide

  • 2021.1
  • 11/03/2021
  • Public
Contents

Monitor Deadline Violations Only

You can run the measurement analysis sample to monitor latency measurements collected from a real-time application, and print only the measurements that exceed the deadline. The real-time application must be instrumented with measurement library APIs.
In this mode, the analysis sample runs the instrumented application, attaches to the shared memory buffer, and starts reading the latency data that the instrumented application placed there. The analysis sample compares the data to the deadline and prints the measurements that exceeded the deadline. After the application is finished, the analysis sample also stops. This type of analysis is called “online” monitoring. This means that the results are constantly read from the shared memory buffer where they are placed by the instrumented application.
The analysis sample monitors the specified measurement instances from the real-time application. A
measurement instance
refers to the measurement library APIs that you have added to your application to measure the latency of the real-time cycle or any part of it. For more information about instrumenting your code with measurement library APIs, see Instrument the Code.
The following example monitors data collected from the multiple measurements sample (
tcc_multiple_measurements_sample
).
To run this example:
  1. From your host system, connect to the target system:
    ssh <user>@<target>
  2. In the SSH session
    , run the sample:
    tcc_measurement_analysis_sample monitor "Multiplication:10:3500" -deadline-only -time-units us -- tcc_multiple_measurements_sample --approximation 10000 --multiplication 100 --iterations 5
    where:
    Argument
    Description
    Multiplication:10:3500
    This argument specifies the measurement instance to collect:
    • “Multiplication” is the measurement name defined in the
      __itt_string_handle_create
      call in the
      tcc_multiple_measurements_sample
      .
    • “10” is the maximum number of measurements that can be stored for this measurement instance.
    • “3500” is the deadline in the time unit specified in
      -time-units
      .
    This argument has the same format as the
    TCC_MEASUREMENTS_BUFFERS
    environment variable. See Control Data Collection.
    -time-units us
    The output will display results in microseconds, and the deadline is specified in microseconds
    -deadline-only
    Print only measurements that exceed the deadline
    --
    Separates the measurement analysis sample command from the real-time application command. For more information about the command-line options of
    tcc_multiple_measurements_sample
    , see Command-Line Options.
  3. Feel free to edit the deadline value to see more or fewer deadline violations. For example, if the previous step did not generate any deadline violations, you can lower the deadline to try to generate deadline violations.
Example Output
The sample prints the output of the profiled application and the profiling results.
The profiling output includes the following:
  • Table of measurement results that exceeded the deadline. Each row represents one measured value. Of the 5 measurements in this example, 1 exceeded the deadline.
  • Summary for each measurement instance:
    • Count of read data: Number of measurements that the analysis sample read from the shared memory buffer
    • Deadline violations: Number of measurements that exceeded the deadline
MONITORING OUTPUT: Monitoring "Multiplication" with deadline 3500 us. ---------------------------------- Name Latency(us) ---------------------------------- Multiplication 3799 Latency exceeding deadline [Multiplication] Count of read data: 5 (deadline violations: 1) ------------------------------ APPLICATION OUTPUT: Running with arguments: approximation = 10000, multiplication = 100, iterations = 5 Running workloads. This may take a while, depending on iteration values. Workloads were run successfully.

Product and Performance Information

1

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