SAP : 3-Tier Architecture:

Application Server/Layer/Tier:- 
It is introduced between client and server to address the above Issues (Two Tier Architecture)

1. The User submit the request using SAPGUI using application server Host Name and Instance number (Port)
2. The requests are received by the dispatcher and kept in the wait queue.
3. Dispatcher maintains the work processes and identifies the load handled by Processes.
4. The dispatcher allocates a freely available process to the user request based on FIFO(First In First Out)
5. The work process contains inbuilt task handlers to interpret the request. The user context rolled into WP(Work process) task handler.
6. The screen in the user request is interpreted by screen handler (interpreter)
7. The Programming language in the user request is Interpreted by ABAP Handler (Interpreter)
8. The SQL Scripts (Commands) in the user request are interpreted by SQL Handler (Interpreter)
9. The request checks whether the content is available in R3Buffer. R3Buffer is used to store the most frequently used content and less frequently modified(updated) content.
10. If the response is not available in R3Buffer then the process goes to the database using DB Specific client and SAP Kernal (OS and DB Specific)
11. Database provides the response to the SAP work process after necessary interpretations.
12. The SAP work process checks whether the response (Content) is eligible for buffering, so that a copy can be kept in R3Buffer for future similar requests.
13. The work process roll out the user related information into user context.                                             User Context: It is a small memory area where the user related information is stored. It contains user authorisations, parameters, common user attributes etc..                                        R3Buffer builds when the instance (application server) is started and R3Buffer is lost when the application server stopped.
14. The response is sent to user.                                                                                                             
• Roll-in: It is the process of copying the user context into WP task handler. When user is logging for the first time the Roll-in is not available.
• Roll-Out: It is the process of copying the user specific information into user context while sending the response to the user by a process. 
• Kernal: Set of executable, binaries and libraries to communicate with OS, DB and Hardware.
• DB Client: Set of executable, binaries and libraries to communicate with the database in the native languages.

Port is derived from a two digit instance number where SAP application server is installed.

Instance number is a two digit ID which ranges between 00 - 99 (where 97, 98 & 99 are reserved)

The Host names and ports are resolved either from /etc/hosts and /etc/services. Most of the companies uses DNS Server for name resolution.

Protocol: It is used to monitor the data transmission between the systems tcp/ip is a base protocol which is used to derive further protocols. Set of rules that Governs transmission of data.

TCP  - Transmission Control Protocol

FTP – File Transfer Protocol

SMTP - Simple Mail Transfer Protocol

HTTP – Hyper Text Transfer Protocol

HTTPS   - Secured Hyper Text Transfer Protocol (using SSL)

SOAP - Simple Object Access Protocol

WAP - Wireless Application Protocol

Port: It is a terminating/accessing/entry point to access the software/device/application.

http :  80/800

https : 443

SMTP : 25

ftp : 21 

The port number ranges between 0-65535. The application reserves the port number to avoid the port conflict. The ports and protocols are maintained in file/etc/services

Server: The system with high availability peripherals to serve the users is called a server. The operating system distinguishes between desktop and server version.

This server version provides a robust, scalable, consistent and reliable platform to host the applications, data bases etc.

Domain server: Used to host the domain in the company.

File server: It is used to provide file storage services in the company.

Mail server: Used to provide mail services for communication from internal and external users.

FTP server: Used to upload the files.

DHCP server: Used to release and renew IP addresses for all the systems that log on to the domain.

Web server: Provides web services in a company.

Print server: Used to host all the printers and provide collaborative print facilities to the users.

Client: It is a system which is used by end users to communicate with the servers.

Client can be a desktop/laptop/webtop/ device with installed software to communicate with server.

Ex: MS Outlook, outlook Express, internet Explorer, lotus notes, SAPGUI etc.

Client is software which contains a set of commands programs, executables, libraries that are used to communicate with servers. He is also referred as frontend to communicate with backend. Client requests and a server Responds. Client is referred as it service requester and server is referred as request processor. In a client server communication the clients are overloaded along with servers due to the following reasons.

1. There is no queue mechanism to process the request. The server has to handle the queue explicitly.
2. There's no inbuilt interpreters (client and server has to interpret separately). No intermediate interpreters.
3. There are no intermediate buffers/cache to reduce hits on the server. Every request hits server.
4. Every client is required DB client to communicate with server. There is a need to deploy and intermediate layer to address the above issues.

The above architecture is also referred as two tier architecture.

Introduction to Host name & IP

Host name: Which system in the data centre is identified by unique name which is called as host name without host name the system could not be communicated over Network.

IP address: Internet protocol address: It is a numeric cable assigned to each machine that is going to participate in the network communication.

Private IP addresses: These are used internally within the company. It ranges between 0-255, 0-255, 0-255, 0-255. (4 bit).

A class IP address starts with 10.0.0.1 (Big companies uses this range)

B class IP address starts with 172.17.0.1 (Middle sized companies uses this range)

C class IP addresses starts with 192.168.0.1 (Small sized companies mostly R&D)

Other than the above all the address are public IP addresses. These IP addresses are obtained from network carriers.

Static IP address: The addresses that is assigned to servers permanently so that addressing to the server is constant.

Dynamic IP address: The addresses that are assigned dynamically which can be renewed for every 14 days based on DHCP(Dynamic Host Configuration Protocol). These are assigned to all desktops/ laptops/ workstation/Devices.

Intranet: The computers which are connected outside the company network using MAN (Metro Area Network)/WAN (Wide Area Network) using public IP address.

Router: It is a physical device that is in entry/exit point to the network. Router can be set with predefined rules.

Firewall: It is a Software/Hardware/Server that is used to restrict/prohibit the access to websites, IP addresses, host names, ports and protocols. 

Various types of Operating systems

Operating Systems are various types:

GUI (Graphical User Interface): Most of the operating systems are windows based and currently Linux operating systems like SUSE/RHEL/CENTOS/OLES.

               

CLI (Command Line interface): CLI are more secured operating systems as most of the viruses are designed for executables. CLI OS version works based on Libraries and Scripts. AIX, HP - Unix and Oracles SUN Solaris.

               

haracter based Operating systems: These are also command line interface operating systems but worked based on characters and numbers. Easy to operate and navigate to the system.

 Ex: AS400/OS400

Memory

Memory: It is used to perform the calculation/data manipulation. It is temporary work area. The data on the memory will be lost when is system is powered off. Volatile memory is computer storage that only maintains its data while the device is powered. Mostly RAM is used for primary storage in personal computer which is a volatile memory. To provide high availability the memory is inserted into different slots. (don't buy 32 GB RAM sticks instead 4 sticks of 8 GB is recommended). Memory is measured in megabytes(MB), gigabytes(GB).

Computer storage:

The data is stored in BITS and bytes. (bits are 0 & 1)

8 bits = 1 byte

1024 bytes = 1 KB (Kilobyte)

1024 KB = 1 MB (Megabyte)

1024 MB = 1 TB (Terabyte)

1024 TB = 1 PB (Petabyte)

1024 PB = 1 Exabyte

1024 Exabytes = 1 Zettabyte 
1024 Zettabytes = 1 Yottabyte 
1024 Yottabytes = 1 Brontobyte
1024 Brontobytes = 1 Geopbyte

Storage: Storage it is used to store the data consistently and it is non volatile. It is measured in GB, TB etc.  Disc performance depends upon the disc speed rpm (revolutions/rotations per minute). 5600/7200/10000/15000 rpm. Servers generally use 15000 rpm and laptops uses 5600 rpm.

Hard disc of various types: IDE, SCSI, SATA, SAS.

To get more performance (IOPS - input output per second) (eye OPS) by more than a single disc.

Ex: 1TB disc is not recommended instead 300 GB x 4 are recommended which will be also used for fail over. 

Network cord: It is a gateway for the computer to communicate with external world. It is used to provide connectivity with external systems. It would be either LOCAL, WIDE, METRO AREA NETWORK (LAN/WAN/MAN) to provide high availability more than one network cord is installed.

Each network cord he is unique and contains its own MAC (Media Access Control) address.

System: The collaboration of CPU, memory, disc and Ethernet code plugged on to your mother board is referred as a system.

The system is identified by its unique name referred as host name. The hostname details are maintained in a file/etc/hosts. The system is hardware which requires software to operate it.

Operating system: It is a set of commands, libraries, executables which are programmed in C, C++,VC++ and java etc.

Data transfer techniques: SAP supports various data transfer techniques to transfer the legacy, parallel processing data into SAP System. It support various interfaces to communicate with SAP and Non-SAP System.

Continuous Support: SAP provides continuous improvement to existing software by releasing patches, packages, support stacks, upgrades and enhancement packages etc.

Best practices / Business Configuration sets: SAP provides are configured scenarios, solutions, building blocks and best practices to accelerate the implementation of SAP. Global best practices reduced cost, risk time.

High availability: It supports all the high availability services of cluster / backup / restore / recovery / standby and disaster recovery solutions.

Optimal response time: It provides optimal response time of 600millisonces to 2 seconds for each step (request through a mouse – click/keyboard stroke like f1, f2, f5, f8 and Enter.

Business Innovation: SAP provides my SAP Business suite business innovations 2011 which comprises of SAP ERP (SAP enterprise resource planning), SAP SRM SAP suppliers relationship management), SCM (supply chain management).

Netviewer: SAP provides a common platform netweaver to install all the required components in a single system (enterprise portal, exchange infrastructure, business intelligence, mobile infrastructure development. Infrastructure can be installed along with ERP system at one step).

Risks:

1. OS/ DB/SAP compatibility with hardware 
2. OS and Hardware Enterprise 
3. Network bandwidth and data transfer rate 
4. SAP support licensing and compatible software 
5. Onsite and Offshore communication 
6. Existing data centre team support and availability 
7. Take backup and restore capability

Hardware: It is a computer hardware in that facilitate various input and output devices to handle the user request and display or output the response to a monitor or printer.

CPU (Central Processing Unit): It is used to process the request by allowing a dedicated or shared process ID/thread ID to the user. The processor power is measured by using mhz/ghz to provide high availability dual/ multiple CPU all installed for failover and load balancing.

Each vendor CPU speed differs from model to model (DELL, IBM, HP)

• CPU can be monitored on windows by using task manager
• CPU is monitored on Linux by using command "top"
• CPU is monitored in SAP system by using transaction "ST06"
• CPU fetches the data into memory and holds as long as it is committed or rollback 

License Fees:

License Fees: SAP Supports a small, midsized and large enterprises and cut down the license cost to reach the customers (ex: each user costs around 2,00,000 INR which has been reduced to 60,000 INR) based on number of users, location, business and volume of licensing etc.