Database apply session balancing - assure_mimix - 10.0

Assure MIMIX Administrator Reference

Product type
Software
Portfolio
Integrate
Product family
Assure
Product
Assure MIMIX™ Software
Version
10.0
Language
English
Product name
Assure MIMIX
Title
Assure MIMIX Administrator Reference
Copyright
2023
First publish date
1999
ft:lastEdition
2024-05-07
ft:lastPublication
2024-05-07T13:36:02.962500

In each data group, one database apply session, session A, is used for all IFS objects, data areas, and data queues replicated from a user journal. If files as well as journaled IFS objects, data areas, and data queues are replicated by the same data group and the data group is configured for single-threaded database apply processing, apply session A can become overloaded or the configuration may cause MIMIX to distribute files among apply sessions in a way that is not compatible with a need to serialize transactions. You may need to adjust the configured apply session in data group object and file entries to either ensure that files that should be serialized remain in the same apply session or move files to another apply session to manually balance loads. Consider the following:

  • In data groups that use default cooperative processing of files (primarily through the user journal), default settings will ensure that newly created database files are distributed evenly across database apply sessions to avoid overloading one apply session. Data group file entries are assigned to apply sessions when a data group is initially started or when a start request specifies to clear pending entries. Default settings use the target system to determine the database file network relationships when assigning apply sessions. However, if the data group allowed MIMIX to manage target constraints prior to upgrading to MIMIX version 9.0, MIMIX can assign files constrained together with foreign key constraints to different apply sessions. Considerations for using target constraint management are described in MIMIX-managed foreign key constraints on target node.

  • In data groups that use legacy cooperative processing for files, newly created database files are distributed to apply session A by default. In data groups that also replicate IFS objects, data areas or data queues through the user journal, it may be necessary to change the apply session to which cooperatively processed files are directed when the database files are created to prevent apply session A from becoming overloaded.

  • Logical files and physical files with referential constraints also have apply session requirements to consider. These must be in the same apply session if target constraint management is not used. For more information see Considerations for LF and PF files .