Example: Inbox processing

Showcase, discuss, and inspire with creative America Data Set.
Post Reply
Mitu9900
Posts: 223
Joined: Thu Dec 26, 2024 9:17 am

Example: Inbox processing

Post by Mitu9900 »

As shown in the Partition-Wise Join section above, it can be beneficial to partition two tables by the same value and method. However, this previously required the same key to be present in both tables. However, this is not always the case and a table definition change is often difficult to achieve at a software vendor.

Since the advent of reference partitioning, two tables can be partitioned the same, even if the partition criterion is in only one of the two tables. The prerequisite, however, is that the connection between the two tables can be established via an active foreign key constraint .


In this example I discuss a type of processing for which I have not yet found an optimal solution. This is the only case in this article where the solution argentina telegram screening shown has not been tried out in practice. It is a simulation on my own database. The basic idea is that a clean room strategy is transferred from one table to another table using reference partitioning.

First, a few words about the topic of the inbox in general. This task occurs again and again in practice and reflects a fundamental problem of the division of labor. Processors get their next task from an inbox (my definition). Such inboxes can appear in different places in applications. Examples are call centers (next call), credit assessment and approval, checking insurance claims, order approval in production and much more.

The challenging thing about finding the next task to be completed is that several search criteria that are located in different tables must be combined. Such search criteria can be, for example:

The status of the order.
The knowledge required to process the order (e.g. by assigning it to a specialised processing group).
The priority of the task.
Below you can see a simplified example of an inbox. An important simplification is that the search is carried out directly for a user number and not for a specialist group. You will now be guided through the solution step by step. Let's start with the basic data structure. The example is about a call center. On the one hand, there is the CALL table, in which the calls are listed in the order in which they arrive. The ASSIGNMENT table shows which calls a specific user could answer.
Post Reply