Push Scheduling Recommendations
Push jobs are run using two main methods, the first of which is manually executing the push job and the second is scheduling the push job.
All push jobs can potentially produce a strain on the UCMDB and SM systems; HPE recommends that you adhere to the following guidelines.
Scheduler time frames
It is important for you to understand the function of the Scheduler “time frame” concept. Running push jobs creates an increase in system activity and may affect application responsiveness. In order to enable users to effectively interact with applications HPE recommends the following guidelines:
In order to reduce system strain, schedule the UCMDB to SM push to run at non-peak usage hours, preferably when system usage is at a minimum.
Scheduler frequency
It is important to be aware of the business requirements when configuring the schedule frequency. The scheduler frequency depends on infrastructure environment changes that must be synchronized between UCMDB and SM.
Define the scheduling frequency based on the business requirements for consuming up-to-date CI information. Most implementations require a daily update. When scheduling small IT systems that are prone to frequent changes, the scheduling frequency may need to be increased.
Push Job dependencies
UCMDB Push Jobs do not support dependencies between each other. Each “Push Job” is considered a separate task and users cannot define job dependencies. For example, that one job is dependent on another or upon completion before the next job is run.
It is important that both CI queries and their dependent Relationship queries exist in the same Job in order to avoid relationships not being pushed to Service Manager. You can change the position of each query in the list to define an appropriate execution sequence. See the following figure for an example.
