To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.

PRoM calculation speed/performance

I'm relatively new to the field of process mining so I'm wondering if the calculation time for PRoM (inductive visual miner) is normally faster than I experience. I have an event log containing 5 columns and +- 75.000 rows.  Column A & B, which I combine in the import as the case ID, column C: activity, column D & E: Start and End time. 
When I load this dataset into the inductive visual miner, the process map is calculated pretty fast, but PRoM then keeps "aligning log with model" for about 10-15 minutes. Once that is complete the metrics become visible (sojourn etc.)

My laptop is a Dell XPS13, 8GB RAM with i7-5500 2.40 GHz

The dataset I'm now analyzing is only a fraction of the dataset I want to analyze in the future. Any remarks on whether it's my laptop or that the dataset is just to large? (or both)

Comments

  • Hi Ruud,

    What you could try is to open the ProM package manager (unless using ProM lite) and on the bottom to increase the memory ProM is allowed to take. On my hardware ProM lite automatically set the max memory to 4Gb (I have 16 Gb), in your case I would set it to 4Gb. You can check in the 'ProMLite11.l4j.ini' file in your ProM lite install directory.

    How long it takes to discover a process model, and then to align the data on top of it, depends on many factors (for instance complexity of the discovered model, how many unique traces there are, etc.).

    In short: sometimes indeed it can take 15 minutes :) sometimes, with similar sized data, its done within a minute...

    Joos Buijs

    Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
    Previously Assistant Professor in Process Mining at Eindhoven University of Technology
  • Thanks, I will check it my settings!
Sign In or Register to comment.