Time Slicing job slow performance

Products

Clarity PPM On Premise Clarity PPM SaaS

Issue/Introduction

The Time Slicing job seems to have increased in the amount of time to finish processing. When running queries to count the number of records yet to be sliced, you see the counts are moving along. You may notice that after the job completes, the next time it starts; there is a large number of allocation records waiting to be sliced again. Why is it taking longer for the job to finish? What are some known issues / steps you can check on what is causing this?

Environment

Release: All
Component: Clarity Time Slicing

Resolution

Check Time Slice durations

Check to ensure that time slicing durations follow the best practices mentioned at: Time Slice configuration best practices

Check for recent changes

Check for recent updates such as the below that can impact how long the job needs to run.

XOG Write actions

Was there a XOG Write action to add many projects with allocations and assignments? If so, new data will need to be sliced for various existing time slice definitions.

Changes in the Clarity Base Calendar

Was there a change in the Clarity base calendar? The Base Calendar controls resource availability and this is taken into the calculation for the allocation and assignment slices. If this calendar is changed, the data will be triggered for a re-slice, and depending on the amount of data to be re-sliced, this can cause the job to run longer.

Changes on an investment team page

Was there a change on an investment team page to the start/finish dates? This will cause the project plan to realign and force the job the reprocess the changes to reflect the change. When users change the allocation start and/or finish date from the team staff page, this will cause a reslice for every member on this team. A better solution would be to change the allocation from the Team > Resource > Allocations tab rather than the Team > Staff page as this will not cause a project date realignment.

Rollover or slice definition updates

Did a rollover occur or were slice definitions updated? A rollover occurs at regular intervals based on the Time Slice definition. There are weekly and monthly roll-over intervals. The Expiration Date on the Time Slice definition will indicate when the next 'rollover' is to occur. Editing a Time Slice definition will erase the data for that slice; however, the data will be replaced when the Time Slicing job is next run. The date range of the slice definition and the amount of data that needs to be re-sliced will impact the amount of time required for the Time Slicing job to finish. If you are modifying Time Slice definitions, be sure to only update one; let the job run to completion and then update another slice definition. This will help to minimize the amount of time required for the job to complete.

Check/monitor the database (DBA)

Have the DBA monitor to see if there are was any contention around the time of the occurrence. There could be other processes running that is competing with the Time Slicing job. GCP customers, please contact Broadcom support for assistance with this step.

Check Known Issues

Below are known issues related to Time slicing performance. You can review this list to see if you are impacted by any of these.

Time Slicing job very slow during monthly rollover (Oracle)

How to speed up the job

Note that once the job is running there is not much that can be done to speed it's progress.
Cancelling and re-running the job will not make things faster as there is still going to be lots of data to process
We recommend leaving the job to complete whilst investigating and addressing the root causes of the slowness as per the above