Informatica has 4 broad layers:

1. client : clients are used to send requests.

Web client:

–Administrator console: we can create services, add or modify nodes, add users and modify permissions and we can perform global monitoring of all activities in informatica.

Standalone client:

–Repository manager: create metadata (ex: creating folder)

Modify metadata (Ex: renaming folder name)

Consume metadata (search)

Repository: it is a storage medium to store metadata

-PC designer: performs design and development of meta elements. Example  create mapping, all the ETL operations are designed here.

–workflow manager:

worksflow means set of tasks to complete bussiness requirement.

(ex: task1 for getting data connected with t2 to load the data).

–workflow monitor: we can perform all the  moonitoring and runtime activities. such as start/run,stop,abort, resume…

command line client:

use this to run and stop workflows , move the meta

– we can schedule the workflows , we use pmcmd command line utility.

-we can also use cmd to move metadata from dev to QA(testing).

2. application server: servers will respond to the requests.

-Tomcat is the default application server that comes with the package.

3. server:

– repository  service: handles the requests made by the repository manager client.
It takes the requests from client tools and stores them in repository service. It maintains connection between client tools and repository database.

Example: Request to create a mapping from designer. When a dev creates a mapping in client. The object details will be stored in repository database.

– integration service: it is used for data movement activities, ex: it moves the data from task1 to task2.

-it will execute the tasks or mappings by accessing the data(metadata) from repository database (object details). Starts the data flow as defined in the mapping.

Repository service is primary service. Integration service depends on repository service.
If a request from client ttol (monitor) is sent to run workflow. It has to get the repository database frow f details which is given by repository service when resquested through client tool.Then integration serice will stat data flow from source to target with transformations.

4. data/meta layer:

Repository database: (technical data) -For metadata storage we have repository database. Which is different from source & targets data.

Source target storage:

-It is the storage media, stores the the output of the ETL.

ex: stores source, target, ods data.

connection btw all the layers:

ex: to create metadata, follows the following steps

powercenter repository manager(requestor)—>create metadata–> appserver–> server–>repository service==>metadata store(Repository)

ex2: create mapping follows:

powercenter designer–>create mapping–>appserver–>server–>repository service–>metadata store.(pc repository)

ex3: start workflow follows:

workflow manager/monitor–>start workflow–>appserver–>server–>(data transfer happens) integration service–>repository service–>datastorage media.

ex4: start workflow with cmd

command line–>start workflow–>appserver–>server–>(data transfer happens) integration service–>repository service–>datastorage media.

ex5: administrator operations(create new service)

administrator console–>create service–>app server–>server–>repository service–>metadata[domain database] not repository database.

Mapping: It allows us to build a ETL flow.