TSCO Aging Configuration parameters description and typical settings

Version 9
    Share This:

    This document contains official content from the BMC Software Knowledge Base. It is automatically updated when the knowledge article is modified.


    PRODUCT:

    TrueSight Capacity Optimization


    COMPONENT:

    Capacity Optimization


    APPLIES TO:

    CO 9.x and 10.0 ,TSCO 11.x



    PROBLEM:

     

    There are 16 tables in the ageing configuration.  What are the typical range set in the field for the various aging parameter settings?  Are there any best practice recommendations associated with the various aging configuration values in relation to data granularity and data retention?

     

     


    SOLUTION:

     

    Legacy ID:KA351802

      

    Here is a short description of the tables available under the aging configuration.

    *_CONF_STAGE: are the tables where the ETLs load data.
    *_DATA_DETAIL: are the tables with the more detailed resolutions
    *_DATA_DETAIL_SPLIT: are an internal tables used by near real time warehouse. Not involved in analyses, models, ...
    *_DATA_DH: these tables contains data at hour resolution
    *_DATA_D: these tables contains data at day resolution
    *_DATA_MDCH: these tables contains month at day resolution
    *_DATA_STAT: not used

      

    In the table below, the 'Usual Aging' field in the table represents a basic 'best practices' range (since that is the typical configuration range).  For the most part the aging period decides how much database space is going to be consumed by the product.  When you are using the Enterprise Partitioning option (which is the recommended best practice configuration) the performance impact of having a lot of historical data is diminished because when you aren't accessing the older the data partitions in which it resides aren't accessed.  When you aren't using Enterprise Partitioning then all of the data is stored in a single region with an index that covers the full period which does have an impact on overall performance.

      

     

      

     

                                                                                                                                                                                                                                                                                                                                                                                                                     

    Table name

    Usage

    Usual Aging

    SYS_CONF_STAGE
     

     

    Configuration data for systems imported by ETLs before warehousing

    Very short like 2/5 days, just to recover from warehousing errors

    SYS_DATA_STAGE

     

    Performance data for systems imported by ETLs before warehousing

    Very short like 2/ 5 days, just to recover from warehousing errors

    SYS_DATA_DETAIL

     

    Performance data for systems at original resolution imported by ETLs

    Short, like 30/60 days

    WKLD_DATA_DETAIL

     

    Performance data for business drivers at at original resolution imported by ETLs

    Short, like 30/60 days

    SYS_DATA_DH

     

    Performance data for systems at 60 minutes resolution

    Medium, like 60/120 days

    WKLD_DATA_DH

     

    Performance data for business drivers at 60 minutes resolution

    Medium, like 60/120 days

    SYS_DATA_D

     

    Performance data for systems at daily resolution

    Long, like 180/365 days

    WKLD_DATA_D

     

    Performance data for business drivers at daily resolution

    Long, like 180/365  days

    SYS_DATA_DETAIL_SPLIT

     

    Performance data for systems at 5 minutes resolution, split for period of competence

    Short, like 30/60 days

    WKLD_DATA_DETAIL_SPLIT

     

    Performance data for business drivers at 5 minutes resolution, split for period of competence

    Short, like 30/60 days

    SYS_DATA_MDCH

     

    Performance data for systems at monthly resolution

    Long, like 1/2/5/10 years

    WKLD_DATA_MDCH

     

    Performance data for business drivers at monthly resolution

    Long, like 1/2/ 5/10 years

    SYS_DATA_STAT

     

    Stats for systems

    Short, like 60/120 days

    WKLD_CONF_STAGE

     

    Configuration data for business drivers imported by ETLs before warehousing

    Very short like 2/ 5 days, just to recover from warehousing errors

    WKLD_DATA_STAT

     

    Stats for workloads

    Short, like 60/120 days

    WKLD_DATA_STAGE

     

    Performance data for business drivers imported by ETLs before warehousing

    Very short like 2/ 5 days, just to recover from warehousing errors

      

     

      

    NOTE:
    There is no “RIGHT” or “WRONG” setting for aging, specific settings depend on the data retention and database space requirements of the environment.  But, in many environments it will make sense to maintain some proportion between the data retention across various summarization levels and keep data longer at the less granularity summarization levels.

      

     Higher resolution (granularity) tables have an higher number of samples and need more space:

      
       
    • detail: has 12 samples per hour per 24 hours per metric per system,
    •  
    • hourly: has 1 sample per hour per 24 hours per metric per system,
    •  
    • daily: has 1 sample per day per metric per system
      

    Also note that it is necessary to reload the warehouse configuration to apply the new settings, and wait for “DB space manager” and “DB cleaner” system tasks to execute to apply the updated settings.

      

    About STAT tables

      
       
    • SYS_DATA_STAT: Statistics for systems.  These are the statistics listed in the "Producer-based statistics" table under Administration -> Data Warehouse -> Custom Stats
    •  
    • SYS_DATA_CUST_STAT: System data tables for Rule-based statistics.  These are the custom statistics that are defined under Administration -> Data Warehouse -> Custom Stats
    •  
    • SYS_DATA_DETAIL_SPLIT: This is typically a small table and it contains data that BCO needs to split across multiple intervals.  So, for example, if you had a 5 minute interval collected from 23:58 - 00:03 then this would appear in this table because BCO would need to split part of it (23:58 - 23:59 to be in day 1 and 00:00 - 00:03 to be in day 2).  This is typically small and thus the aging rules for this table aren't really important for space management.
      
      

     


    Article Number:

    000306000


    Article Type:

    Solutions to a Product Problem



      Looking for additional information?    Search BMC Support  or  Browse Knowledge Articles