Infor LN & Baan Tips & Tricks for TECHNOLOGY: Data Sharing Methods (Advantages and Disadvantages)
Depending on the multicompany scenario you choose, an implementation team must decide whether or not tables must be logically linked or if data must be synchronized in another way to achieve availability of data across various companies.
There are 3 ways in which data may be shared among companies. Here are some advantages and disadvantages of each:
-
Logical Table Linking - If two companies use the same physical data, the physical table exists, or is used, in only one company: the physical company. Each piece occurs only once: one instance of the same data. If the term logical table linking is used, users from multiple companies use a single physical instance of the data. If the company tables are on the same server, this can be accomplished by logical table linking. Access to specific data can be restricted, if required.
-
Advantages: Logical Table Linking takes place in real time; therefore, the moment a record is created or modified in one company, a record becomes available in all other linked companies as well. The setup and maintenance of logical table linking is easier than the setup and maintenance of data replication. Logical table linking is extremely reliable because this type of linking is independent of network connections and user interventions.
-
Disadvantages: Table sharing implies that all attributes for a record are the same in all companies. Therefore, suppose you share the item table and, for a particular item in a company, the product class is XXX. In this case, in all other companies, the product class for this item must be XXX as well.
-
Data Replication - In this situation, each company has exactly the same data, but each company has a unique copy of the data. The same piece of data exists in multiple (physical) places: multiple instances of the same data. The process to copy the data is called data replication.
-
Advantages: Rather than using table sharing by means of the logical linking of tables, you can replicate the content of tables between companies. The advantage is that, on company level, some (non-key) attributes of a record can differ by company. For example, if the bills of materials are replicated instead of shared, for each company, you can link a different warehouse to the bill of material. As a result, the bills of materials can be the same across all companies and only the warehouses differ. Using replication, you can also make only a subset of the records available in other companies. For example, if you replicate items between companies, in a sales company, for example, by means of the item group, you can only make end items available. In addition, you can replicate only a subset of items, for example, depending on the item group. Note that replication also requires that the referenced tables are replicated or shared.
-
Disadvantages: Replication is not in real time, and therefore, for processes that require real-time integrated data as the financial integration processes, replication is not an option. For replication, you also must take into account the sequence in which you replicate. First, you must replicate the child tables and then the parent tables, and therefore first the business partners and then the purchase contracts. During the replication process, the data must be frozen. This point is of particular importance if the replication process takes a lot of time.
-
Manual Syncronization - If only a small number of records are the same between some companies, you can enter, maintain, and keep the records synchronized manually. Note that the more dynamic data is, the more difficult this process is. This solution is strictly procedural. Therefore, depending on the discipline of the users, the solution is error sensitive: data can be forgotten, updated too late, or typing errors may occur.
-
Advantages - The main advantage is the flexibility so that only the data that really must be shared can be kept synchronized. By manually keeping data synchronized, not all attributes of a record need to have the same value.
-
Disadvantages - The main disadvantage is that this method depends on user interaction. Therefore, the method depends on the user’s time - because the method is not a real time procedure and the user may forget to update the data - and users can make mistakes.