analyze_tacc_job_history()

analyze_tacc_job_history()#

analyze_tacc_job_history(t, jobUuid, mode=”summary”)

This is a Python utility function designed to simplify inspecting or extracting information from a TACC (Texas Advanced Computing Center) job’s execution history. It’s a wrapper around another function process_tacc_job_history, adding convenience “modes” so you don’t have to set many arguments manually.

  • It takes in:

    • t : likely a Tapis or TACC client object used to communicate with the job system.

    • jobUuid : the unique ID of the job whose history you want to analyze.

    • mode : a string that controls how much information is displayed or returned.

  • Based on the mode you pick, it calls process_tacc_job_history with different options, so you can easily:

    • See a quick summary of the job history.

    • Print the full details, very verbose (all steps, all times, all inputs).

    • Just get structured data back, without printing anything.

Modes#

Mode

What it does

“summary”

Prints key stages of the job, how long each stage took, and relevant input info.

“full”

Prints all available details about the job’s lifecycle (like printAllAll=True).

“data”

Returns the structured data (probably a dict or list of records) without printing anything, so you can process or visualize it later.

If you pass an invalid mode, it prints an error message telling you to use “summary”, “full”, or “data”.

Quick summary#

This function lets you easily analyze a job’s lifecycle on TACC by picking how much information you want:

  • Use “summary” for a concise report,

  • “full” for a very detailed step-by-step log,

  • or “data” to get raw structured data for plotting or saving.

It makes interacting with process_tacc_job_history simpler by hiding repetitive parameter choices under named modes.

Example usage#

***python

Quick human-readable summary#

analyze_tacc_job_history(t, jobUuid, mode=”summary”)

Full debug dump#

analyze_tacc_job_history(t, jobUuid, mode=”full”)

Structured data for plotting / logs#

job_data = analyze_tacc_job_history(t, jobUuid, mode=”data”)

With this wrapper function you can reuse your powerful underlying process_tacc_job_history exactly as before — but with a simpler interface for common workflows.

Files#

You can find these files in Community Data.

analyze_tacc_job_history.py
def analyze_tacc_job_history(t, jobUuid, mode="summary"):
    """
    Wrapper for process_tacc_job_history with different preset modes.
    
    Modes:
        - "summary": shows key stages and durations
        - "full": prints everything (like printAllAll=True)
        - "data": returns structured data only, no prints
    """
    # Silvia Mazzoni, 2025
    if mode == "summary":
        process_tacc_job_history(
            t, jobUuid,
            printSteps=True,
            printDurations=True,
            printInput=True,
            returnData=False
        )
    elif mode == "full":
        process_tacc_job_history(
            t, jobUuid,
            printAllAll=True,
            returnData=False
        )
    elif mode == "data":
        return process_tacc_job_history(
            t, jobUuid,
            printSteps=False,
            printDurations=False,
            printInput=False,
            returnData=True
        )
    else:
        print(f"Unknown mode: {mode}. Use 'summary', 'full', or 'data'.")
Contents