SAP Architecture Components

The SAP R/3 3-tier Architecture is a popular application server topology for developing large-scale
enterprise applications. With the increased use of web-based technologies and the ever-increasing
importance of user experience, application servers are being replaced by web-based applications.
However, there are certain situations where the usage of web-based applications is not possible, due
to the size of the system, regulatory compliance, or other business reasons. In such cases, the usage
of 3-tier application servers is unavoidable. The most common application server topology is the SAP
R/3 3-tier Architecture.

 The role of the Message Server is to handle communication between ABAP Dispatchers that
are distributed across the system.

 In this Dispatcher queue, process work of several different types is stored.

 Work processes are allocated by the dispatcher.

 Gateway provides access between SAP systems and external systems.

 Each R/3 dialog step is separately executed by ABAP-Work processes.

The following describes the types of work processes:

 The Java dispatcher and server processes are handled by the Message Server. Within the Java
runtime environment, the Message Server enables communication.

 A server process is configured by a Java application program to handle logical locks. The
queue server handles them.

 The central services are used to manage locks and transmit messages and data when
working in a Java cluster. Java clusters are sets of cooperating processes that create a
dependable system. A group of resources, such as memory, work processes, and so on, is
known as an instance.

 The Dispatcher handles the incoming client requests and dispatches them to the correct
server process.

 To install J2EE components using SDM, you must first download and install Software
Deployment Manager.

 A high quantity of requests can be handled by a Java Server Process.

 Threading refers to the method of processing multiple items at once in the background.

 SAP can be accessed from a browser by entering the system’s URL. Because of this, ICM can
also be used for communication.

When using the ABAP and Java systems, we use another component called a JCO to handle
communication with the ABAP dispatchers. ABP+Java uses Java dispatchers to handle
communication.
How Does the SAP Process Work?

 The SAP Logon process is the first step of the SAP authorization process. This is a simple
process which involves entering the identification number that you were assigned when you
created an account with SAP.

 After this, you will be asked to provide your SAP password.

 After you successfully enter your SAP password, you will be sent a confirmation message
confirming your successful logon. From here, you will be able to access the main menu of
SAP.

Although the process may seem simple at first glance, it can sometimes be challenging to perform
correctly. Therefore, it is recommended that users practice logging in before conducting business-
critical activities such as signing contracts or performing transactions.

 The Dispatcher listens to user actions such as clicks on links or forms and forwards them to
corresponding actions. When the action receives the request, it handles the request and
returns the response to the Dispatcher. The Dispatcher forwards the response to the system.
 Once a process is assigned, the dispatcher will check if the resource is available in the Service
Request Pool. If it is not, the dispatcher will go ahead and start the process and assign
another work process when the work process is available. If the resource is available in the
Service Request Pool, the dispatcher will find another process and assign that one and so on.

 The work process is responsible for putting work into the system and delegating it to other
processes. Processes can be divided into two categories depending on their priority. These
are critical processes which need to be executed immediately. These processes are called
high priority. Critical processes and other high-priority processes can be assigned to users.
Thus, depending upon user request work process is assigned.

 In the process of work, if a user does some changes in the process and the process is not
completed, then the User will get an alert message. In the case of the dialogue step, when a
user authorizes the work process, then shared memory will be checked. If any other user’s
data is present in shared memory then the process will be stopped and an alert will be
displayed.

 Another work process is reading. If there is no data in the buffer then data has to be
retrieved from the database to get a response. And this process is called reading. If there is
data in the buffer then it has to be removed. And this process is called kill. This process
increases the performance of the system. It is very important to keep these work processes
in mind while designing your application.

 A dispatcher is a class that receives requests from the GUI and sends the requests to the
server. The server receives the request and executes the query and returns the result to the
dispatcher which then updates the GUI. A dispatcher can be registered at the startup of the
application and it can be unregistered at any time to stop the dispatching. By using this
approach, the data is synchronized between the server and the GUI.

 If the shared data is not being used by any other user, it will eventually be evicted from
shared memory and the system will have less memory to service other workloads. It is
important to monitor data in shared memory for signs of stagnation, because once data is in
shared memory it cannot be purged.