nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.

How to find process bottleneck?

gromar

Hello all.

How to find process bottleneck?

I tried to answer this question. My answer was to generate process diagram with highlighted nodes where tasks (or executions) are gathered.
[img]https://raw.github.com/gro-mar/activiti-crystalball/master/simulator/src/test/resources/org/activiti/crystalball/generator/ProcessInstanceMonitor-1.png[/img]

There are 4 tasks in the User Tasks 1. It seems that User Task 1 assignee will have a lot of work.
To read little bit more visit link.

I would appreciate any feedback.

Find more posts tagged with

Comments

jbarrez

Yes, I understand the approach you take. It's simple and that's why I also like it.

However, as a sidenote, to take in account that Activiti will create executions on the fly sometimes to eg. create parallel steps, subprocesses, etc.
In that case, you would have for example an execution in the parallel gateway split, which actually doesn't really is a bottleneck.

So you'd need some specific logic to remove those again, or only focus on user tasks. Most of the times, the 'business bottleneck' will be the human actors, not the system.

gromar

However, as a sidenote, to take in account that Activiti will create executions on the fly sometimes to eg. create parallel steps, subprocesses, etc.
In that case, you would have for example an execution in the parallel gateway split, which actually doesn't really is a bottleneck.

I would like to learn something more. Could you explain it little bit more, please?
Why it does not work for parallel steps?

jbarrez

because the way Activiti is implemented under the covers. The easiest to learn is to write some unit tests, debug them and check the database contents as described in http://activiti.org/userguide/index.html#apiDebuggingUnitTest. Especially the execution table.

For the parallel gateway, when it forks, one execution is kept as 'inactive parent'. For all the outgoing sequence flow a new child execution is created. Hence my remark, that you'd have many executions pointing to the parallel gateway, until the corresponding join is entered. That would give you false positives. But like I said, nothing you can't fix by knowing how the executions work.

gromar

You were right.

I changed reporting little bit.
[img]https://raw.github.com/gro-mar/activiti-crystalball/master/simulator/src/test/resources/org/activiti/crystalball/simulator/simulateBottleneckTest-2.png[/img]

I have an issue with adding new users to the process engine (ACT-1543).
Is it valid issue? (Or I did something wrong.)

frederikherema1

You should save the user, prior to creating the membership. The JavaDoc on newUser clearly states this:

/**<br>&nbsp;&nbsp; * Creates a new user. The user is transient and must be saved using <br>&nbsp;&nbsp; * {@link #saveUser(User)}.<br>&nbsp;&nbsp; * @param userId id for the new user, cannot be null.<br>&nbsp;&nbsp; */<br>&nbsp; User newUser(String userId);

gromar

I am sorry. It was my fault.

Thank you for your answer.