ascp data collection technical workshop good
Post on 28-Nov-2015
Embed Size (px)
DESCRIPTIONASCP Data Collection Technical Workshop.
APSAPSAPSAPS---- Data Collection Data Collection Data Collection Data Collection 1111////14141414
Data Collection Technical Workshop Sanjeev Kale Oracle Corporation Introduction This paper focuses on understanding Data collection architecture. Data Collection is a process that pulls data from designated data sources into its planning data store. The Data collection process consists of the pull process and Operational Data Store (ODS) load process. Note: This document is intended for use as a reference document. Certain sections of the text are intentionally repeated throughout the document, so that each section can be read and understood independently of one another. This document is designed to supplement Oracle Advanced Supply Chain Planning and Oracle Global ATP Server Users Guide and other APS training class notes. Overview of Running Collections Data Collection programs move data from the source instance into staging tables where data integrity is checked before being loaded into APS instances Operational Data store. Pull Data programs move the data from the staging tables to the APS instances Operational Data store. The user has the flexibility in determining when the snapshot of information from the transaction system (Source Instance) should be taken, and in deciding what information to capture with each run of Data Collection. The data collection program can be set to run upon submission of a job request, and at specified time intervals, and to collect different type of information with different frequency e.g. Dynamic data such as sales order can be collected frequently, while static data such as department resources can be collected at longer intervals.
Definitions Oracle Applications Data Store (ADS) Represents all the source data tables used to build and maintain the planning data store within Oracle ASCP. It represents a single source data instance. E.g. transaction system (Source Instance) Operational Data Store (ODS) Part of Oracle ASCP that represents all the planning data tables that act as destinations for the collected data from each of the data sources (both ADS and Legacy). This acts as the input for the snapshot portion of the planning process. ODS and PDS share the same physical tables where a special plan identifier (for example, -1) is used for distinction. Planning Data Store (PDS) Represents all the tables within Oracle ASCP which encompass those in the ODS and other output tables from planning, including copies/snapshots of the input data used for the current planning run striped by the appropriate plan identifier. Data Collection Data collection consists of the following:
APSAPSAPSAPS---- Data Collection Data Collection Data Collection Data Collection 2222////14141414
Pull program Collects the data from the ADS, and stores the data into the staging tables. This pull program is a registered AOL concurrent program that could be scheduled and launched by a system administrator. If you are using a legacy program, you must write your own pull program. The pull program performs the following major processes.
Refresh snashots. Launch the pull Workers to perform
pulls from the appropriate source tables and insert the data in the Staging tables.
ODS Load A PL/SQL program which performs the data transform and moves the data from the staging tables to the ODS. This collection program is a registered AOL concurrent program that could be scheduled and launched by the system administrator. The Launch_Monitor procedure performs the following major processes: Key Transformation - Generate new
local ID for global attributes such as items, category set, vendor, vendor site, customer and customer site.
Launch the ODS Load Workers to perform Create, Update, and Delete operations for each entity in ODS (MSCPDCW).
Recalculate the sourcing history based on the latest sourcing information and the data from transaction systems.
Recalculate the net resource availability based on the calendars, shifts and department resources information.
Purge the data in the Staging tables (MSCPDCP).
Collection Workbench Collection Workbench is a centralized data repository providing collected data from the source. Data from different source instances can be viewed using the Collection Workbench. The functionality here is similar to the Planner Workbench functionality. The collection
workbench is used to verify that the intended data has been collected If necessary, troubleshoot errors in data collection and rerun the data collection program. Data Collection Setup Steps The Data Collection can be implemented in two configurations Centralized configuration and Decentralized or Distributed configuration. Centralized configuration is where the transaction (Source) instance and destination (APS) instance is on the same instance. Decentralized configuration is where the transaction (Source) instance and Destination (APS) instance are two different instances. These two instances could be on different machines. Depending on the type of configuration that is selected, the setup steps will change. E.g. For Centralized configuration there is no need to set up database links, whereas in Decentralized configuration, the database links will have to be set up before proceeding with any other setup steps.
The Oracle APS setup for data collection involves three steps Establish database links Define Instances / Associate
Organizations from where the data is being pulled.
Specify parameters for the data to be collected from each instances for each named request set
Database Links The Planning Database Link is defined by the database administrator on the APS (destination) planning instance. There are two Database links created. One on the source instance that points to the destination instance (known as the Application Database Link). One on the destination instance that points to the
APSAPSAPSAPS---- Data Collection Data Collection Data Collection Data Collection 3333////14141414
source instance (known as the Planning Database Link). Both the links are bidirectional. The Planning Database link is used for Data collection from transaction source
to the planning instance. When an action, such as releasing a
planned order or requisition, occurs in the APS planning system, the data is published to the transaction instance. The first step in this process is to send a signal via the Planning database link to the transaction instance. This initiates a remote procedure that pulls the planned order or requisition from the planning instance to the transaction instance.
The Application Database link is used for Completing the Publishing process. The remote procedure that pulls the
planned order or requisition record from the planning instance to the transaction instance does so by using the Application Database link.
Centralized vs. Distribution Configuration
This refers to the deployment of planning and ERP modules.
In Centralized deployment, planning and ERP modules are in one machine and in the same database instance.
In the Distributed configuration, planning and ERP modules are on different database instances and are usually on different machines.
Centralized vs. Decentralized Planning
This has nothing to do with the way the planning and ERP modules are deployed. This has to do with whether you are planning all orgs in an enterprise together or separately.
In centralized planning, you plan all orgs in ONE PLAN.
In decentralized planning, you plan different subsets of orgs in different
plans and link them up by feeding the supply schedule of one org as demand schedule of another.
Example of a Database link Following database link is created in Source instance: In source (Application Database Link) ============================= create public database link APSconnect to apps identified by appsusing 'APS.world'; Following database link is created in Planning instance: In APS (Planning Database Link) ========================= create public database link vis1153connect to apps identified by appsusing 'vis1153.world'; Test the APS link as follows: In APS SQL*Plus: ==================== select sysdate from dual@vis1153; Test the source link as follows: In vis1153 SQL*Plus: ==================== select sysdate from dual@aps; These are the tnsnames.ora entries needed to connect to both instances: APS.world = (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=aps)(PORT=1523))(CONNECT_DATA=(SID=APS)))
vis1153.world = (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=aps.us.oracle.com)(PORT=7500))(CONNECT_DATA=(SID=vis1153)))
Define Instances Navigation path for defining the instances.
APSAPSAPSAPS---- Data Collection Data Collection Data Collection Data Collection 4444////14141414
Advanced Supply Chain Planner SetupInstances The Database Administrator uses this form to setup instance names and to specify the release version of the source database, Application Database link and Planning Database link associated with the instance names.
Complete the following Field and Flag in the Application Instance Window. Set up Organizations by clicking on the Organizations Button. The Organizations window lists the organization within the instance. To enable data collection for an organization within an instance, select the enabled check box. The database links can be confirmed by running the following SQL scripts after the instances are defined. In vis1153 SQL*Plus ================ Select db_linkfrom dba_db_linkswhere db_link = 'APS';
select m2a_dblink,a2m_dblinkfrom mrp_ap_apps_instances;
MRP_AP_APPS_INSTANCES table points