Searching and investigating the datastore
The Search and Investigate stage of the mapping process is when you use the seed data obtained from the application owner as a starting point to figure out the structure of the application. You perform searches in the datastore based on search terms from the seed data, and use the search results as clues to do more searches until you eventually have the whole structure of the application. As you go along, you add what you find to the prototype application map.
See the following example for a walk-through:
Before you begin
This stage requires certain skills and knowledge on the part of the application mapper. For more information, see the description of the application mapper role.
The premise of this stage in the workflow is that you find the infrastructure that supports the application by searching in the BMC Discovery datastore. This works well because, in general, an application mapping initiative starts when an BMC Discovery rollout is already far advanced. It is likely that the items supporting the application have already been discovered, and they just need to be found in BMC Discovery and brought together in an application map.
Searching for application components
You use what you have learned from the seed data provided by the application owner to begin your investigation. For example, you might find in your research that the Tomcat Software Instance (SI) depends on the database details. You might discover that a Tomcat search links it to certain modules. Other things to look for include:
- Search for terms in your seed data.
- Use what you find to try to understand the type of application it is, and add the items to the prototype, if appropriate.
- Look in the items you have found for more search terms and search for them. For example:
- Database names
- Software names
- Windows service names
- Process names
Next, you gather all this information into subgroups to group application components into more refined blocks that represent the tiers in the application (for example, one for the web tier, and one for the DB tier). This is the beginning of your prototype.
George has responded to Mike's e-mail. He tells Mike the following about the application:
- The name of the application is Friends.
- There is a Web component and a database, but he can't provide any further details.
Beginning with an index search, Mike starts to search the application for all the relevant infrastructure items, and continues his investigation until he can see how the application holds together, or until he needs to ask George more questions.
Where to go from here
As you build the prototype to find relevant infrastructure items, you must structure it to enable enable you to place components into groups that match the structure of the application. This refinement will enable you to get back to the application owner to check that the basic structure and content is correct.