Advanced Peer-to-Peer Networking (APPN) is part of an IBM Systems Network Architecture (SNA). It involves protocols in which a set of computers are not dominated by a central server or any other piece of hardware rather these protocols allow computers to communicate among themselves.
Need of Peer-to-Peer Networking
The advent of the worldwide web caused a new problem. There weren’t enough nodes to support the then developed web-based applications as the incoming traffic saw a steady increase, this resulted in a shortage in the network bandwidth of the client-server setting. Peer to peer systems which were later developed offered a solution where every node of the system would act as both the client and server for a variety of domain applications that would provide part of the overall resources available from the system. Thus there exists no centralized coordination in a pure peer to peer networking system where the participating peers have their own freedom and thus are able to self organize the structure of the system. Thus this peer to peer network addresses many issues raised by the conventional client-server systems but it too has its own set of disadvantages which we will be looking at.
History of APPN
The Systems Network Architecture (SNA) is IBM’s networking architecture developed in 1974 as a means of communication between the mainframe computers. It has a variety of interfaces that permit software and hardware communication. Since it has a predictable and robust response it was used in many applications. However, its inability to provide dynamic connections between the SNA resources and its effect on the end-users session in case of software or hardware damage along the route was its major setback. Thus the APPN has developed whose nodes had the ability to understand and interpret the network topology by identifying the location of various nodes in the local network with the ability to make use of current and adjacent networks to search for resources. Hence APPN identifies and selects the best available route between session partners.
Working of APPN
- As the directory services are distributed, a Node needs to remember only the resources that make use of its services. However, directory services can be centralized using VTAM.
- It dynamically locates and defines resources and routes with the ability to establish sessions between any two logical units of the network, without the need for the requirement of a mainframe.
- As we saw before the APPN has the ability to understand and interpret the network topology and hence establishes the best available route between session partners. In case there are any changes in the network, the topology is updated accordingly.
- In APPN, Class of service actually extends out to the end nodes in the network rather than just between front-end processors as in the case of conventional client-server systems with the ability to define cost and other major features.
Role of the APPN nodes and their functioning
The APPN is composed of clusters of T2.1 nodes linked to each other.they provide sessions between Logical units and peer-level connectivity using the APPC protocol. They perform the following tasks:
- Information about the resource location they own is dynamically exchanged between them.
- It keeps a track of the SNA resource locations within the network
- since it understands and interprets the network topology it selects the best available path between session partners.
Types of T2.1 nodes
- Low-entry networking (LEN)
- APPN end node EN)
- APPN network node (NN)
- Specialized VTAM nodes
Advantages of APPN
- Dynamic definition of available network resources which makes connections, routing and reconfigurations simpler
- Automation of resource registration and directory lookup
- Flexibility, which allows APPN to be used in any type of network topology
- It is easier to set up hence saves the configuration and implementation time
- Ability to isolate terminal failures.
- It is cost-effective
- In peer-to-peer networks all nodes act as a server as well as client, therefore, no need of dedicated or centralized server, hence unlike a conventional client-server network, which can fail if the central server malfunctions. This is not the case here.
- Due to the fact that each client in a peer to peer network is a server, the performance of this network increases when more clients join it, unlike a traditional client-system server where this is not the case.
Disadvantages of APPN
- Files and Folders cannot be centrally backed up
- User performance can be slowed down as each computer can be accessed by many users
- Since Files and resources are stored in individual computers they are not centrally organized and hence can be difficult to locate.
- Since users often don’t need to log onto their workstations there is little or no security apart from the permissions.
- It must be safeguarded against virus attacks else the individual users must pay for the losses they face due to these attacks.