Switching Technique

Introduction

• A switch is used to connect the incoming link to the desired outgoing link and directs the incoming message to the appropriate outgoing link.
• A communication switch is used to allow/stop the flow of messages through the path connecting the receiver and the transmitter. 
• The role of a switch becomes more important when there are a large number of users and a particular user at one time may want to communicate with another user, and wants to communicate with still another user at a different time. Thus, the same user has to be connected to two different users at two different times. This can be done by a controlled switch.

Definition

• The process of making a communication link between sender and receiver using a switch is known as switching.

Features/Characteristics

• In a switched network, the temporary connection is established from one point to another for either the duration of the session (as in circuit switching) or for the transmission of one or more packets of data (packet switching).
• Switching plays a very important role in telecommunication networks.
• It enables any two users to communicate with each other.
• In message and packet switching, the incoming message/packet to the node is stored temporarily in a memory location. Then, the stored message/packet is re-transferred to another desired memory location, from where the message/packet can be delivered/forwarded to the next node or the receiver. The transfer from the incoming bin to the outgoing bin is done with a control/command signal.

Advantages

• Switching is used to connect two nodes that are not directly connected to each other to transfer data between them successfully.

Types of Switching

• There are three categories of switching – 
  (A) Message Switching
  (B) Circuit Switching
  (C) Packet Switching

(A) Message Switching (Outdated)

• • Message switching is one of the initial/earliest modes of switching, which helped a lot in the proliferation of electrical communication, but is now outdated, i.e., not used now.
  • Telegraph/Telegram, which arrived earlier than the Telephone, is one of the best examples of Message Switching.
  • This method of switching works on the principle of ‘Store and Forward’.
  • Unlike Circuit Switching, no circuits are switched/used in this mode of data transmission.
  • Working Mechanism of Telegraph –
    • Initially, the message is taken from the user and sorted first on the basis of the receiver’s address and clubbed with other messages moving in the same direction/destination, then they are bundled and sent.
    • The operator at the next intermediate node receives all these messages, stores, sorts, and forwards them.
    • Here, Information is transmitted through various nodes as discrete messages. Hence, this method of switching is known as Message Switching. Here, the storage & sorting of messages was done manually.
    • Later on, the storage process was automated using paper tapes. The advantage of using paper tapes is that the incoming signal is punched onto it automatically, and the same tape can be directly fed into the telegraph machine for further transmission.
    • As the teleprinters came, Morse code was replaced by machine telegraphy, resulting in faster operations for sending messages. Later, computers were introduced to perform the function of message switching. Computer-based message switching is still used in many organizations with many locations of work.
  • Mechanism of Message Switching –
    • In this technique, at each node (telegraph office), the message that arrives from the previous node in the path is stored for some time, sorted, and depending on the availability of the path from this node to the next in the path, the message is forwarded.
  • Message Switching (telegraph) is less costly than Circuit Switching (Telephone) due to –
    • Better utilization of transmission media.
    • The message switching is done over a distributed time.
    • Hogging (Capturing the entire path) does not occur in message switching. Only
      one of the links in the entire path may be busy at a given time.
  • Demerits –
    • Message Switching requires storage, and this may raise the cost of the systems.
    • This Switching method is not effective for sending multimedia data.

(B) Circuit Switching

Definition :

• • Circuit switching is a switching technique in which a particular (dedicated) physical communication link is selected/created statically between sender and receiver via various intermediate nodes, from available physical links to transmit data/messages successfully.

Features :

• • In this mode, the user is allocated the full use of the communication channel for the duration of the call.
  • Circuit switching involves creating a switched path for the entire communication and is available for the entire communication.
  • This mode of switching is said to be connection-oriented, i.e., when a sender and a receiver wish to communicate with each other, the sender has to first dial the numbers of receiver. Once those numbers are dialed, the originating/sender exchange will find a path to the terminating/destination exchange, which will, in turn, find the sender. Finally, the circuit or channel has been set up, then communication will take place; this mechanism is referred to as being connection-oriented.
  • In this mode of switching, the physical connections are made by the switching offices to connect intermediate switching elements between the sender and receiver to complete the call between two users.
  • For each connection, physical switches are set/linked in the telephone network to create a physical circuit. This is the job of the switching office that links switches and sets up at the beginning of the connection and maintains throughout the transmission period. 
  • Telephone/Telecommunication (PSTN) is one of the best examples of Circuit Switching.

Advantages :

• • • Once the physical circuit has been established, the dedicated path will remain to exist until the connection is terminated, and communication is fast and without error.
    • It is highly reliable.

Disadvantages :

• • • A major drawback of circuit switching is the requirement of a dedicated path between the calling/sender and the called parties/receiver. This dedicated path reserves network resources like the chain of intermediate switches and connected transmission media over the entire path during the communication period of time which is obviously a costly proposition. This is not a very efficient strategy as a connection holds a physical line even during silence periods (when there is nothing to transmit/talk).
    • It involves a lot of overhead during channel setup.
    • It wastes a lot of channel bandwidth, especially when a user is sometimes mainly listening and not talking.
    • It may take a longer time to set up between the sender and receiver.

(C) Packet Switching

Definition :

• • • Packet switching is a switching technique in which a logical communication link (either virtual circuit or datagram circuit) is created dynamically between sender and receiver via intermediate nodes, from available physical links to transmit data/messages in the form of packets successfully.

• • • Packet Switching is the backbone of present-day communication systems.
    • Packet switching is the most favored technique in present-day communication systems.
    • Packet switching works on the principle that long messages are fragmented into small units, known as packets. It is these packets that are transmitted instead of a single long message. These packets are now transmitted over the network in the same manner as the messages in message switching. The packets are stored and forwarded at every node. Every packet has the source and destination addresses. Like message switching, repeated transmission of addresses at every intermediate node in this mode consumes network bandwidth.
    • Packets :
      • A packet contains the address of the computer that sent it and the address of the computer to which it is sent.
      • In general, packets need not be of the same size.
      • The Internet Protocol(IP) specifies the maximum size of packets in the form of a maximum transmission unit (MTU) and does not give the minimum size of packets.
      • Packets are generated by the network hardware.
  • Examples :
    • The Internet is one of the best examples of Packet Switching.
  • Advantages :
    • Packets of the same message are launched into the network in parallel way over different available forward links at a node. These packets would travel through different paths to arrive at the destination. This simultaneous transmission of packets over different paths results in further improvement of the link utilization compared to message switching.
    • No link/channel is engaged for a long time since the size of the packets is smaller than the single message, as in message switching.
    • This switching permits better sharing of the links/channels amongst multiple users.
    • Packet switching is cost-effective.
  • Disadvantages :
    • In packet switching, the overhead/wastage is more because every packet is required to carry the addresses on its head/header, which are decoded each time on every intermediate node.
    • The network bandwidth consumed is the maximum in packet switching and the minimum in circuit switching.
    • Firstly, the packets of the same message traveling through different paths may arrive at the destination at different times due to different delays encountered in different paths. Thus, the packets may arrive out of order. In order to deliver them to the destination, they need to be ordered, which requires extra processing and so more delay. They need to be given sequence numbers
      for reordering them. The sequence number increases the overhead and requires more network bandwidth.
    • Secondly, some of the paths may not be very good, and some packets may get lost. This worsens the quality. To improve quality, they require re-transmission, which in turn requires more processing time and more bandwidth.
  • Types of Packet Switching :

Packet Switching occurs in two ways: –

(a) Connection-Oriented/Virtual Circuit Packet Switching and (b) Connectionless/Datagram Packet Switching

• • • Virtual circuit data transmission provides a connection-oriented service over packet-switched networks, ensuring reliable and ordered delivery of data by establishing a logical path before data transfer.
    • In computer networks, a virtual circuit is a communication method that establishes a logical connection (using from existing physical connection) between two nodes (normally the source and destination) before data transmission begins.
    • In this mode of transmission, all data packets from the same source going to the same destination are transmitted over the same routes/channels and through the same routes during the session.
    • It is widely used in networks where quality of service and reliability are important.
    • This transmission results in almost a constant delay for the different packets, and the different packets reach their destination in order.
    • In this mode of transmission, packets arrive in the order they were sent (Sequential delivery).
    • Virtual circuit may be of two types: –
      • Permanent Virtual Circuit (PVC): This virtual circuit is pre-configured/pre-established by the Administrator and is for long-term connection (e.g., leased lines). This connection is always available, even if not in use.
      • Switched Virtual Circuit (SVC): This virtual circuit is created and removed dynamically for each session (e.g., ATM, X.25). They are used for temporary connections.
    • This process of transmission is also called Virtual-Circuit Packet Switching, as it involves the establishment of a fixed logical path from the available physical paths for a talking period of time, which is called a Virtual Circuit or Virtual Connection between the source and destination before the transfer of data packets.
    • The virtual circuit creation takes the following steps:-
      Step 1: Connection Request
      Step 2: Connection Confirm/Establishment 
      Step 3: Data Transmission Start
      Step 4: Connection Release/Terminate
    • Examples: ATM networks, X.25 networks, Frame relay transmission, MPLS (Multi-Protocol Label Switching). 
    • Advantages:
      • They have less packet loss.
      • They face comparatively less congestion.
      • They perform reliable data delivery because packets arrive in order.
      • This transmission guarantees bandwidth and quality of service to the data packets (Resource Reservation).
      • In this mode of transmission, data is sent in packets, each carrying a virtual circuit identifier (VCI) instead of a full destination address. Only the VCI is needed for forwarding, reducing overhead. Thus, it supports a simple routing process.
      • They simplified the error handling process (single acknowledgment for the entire circuit).
    • Disadvantages:
      • Initial connection establishment takes time (Setup delay).
      • Less suitable for a congested system.
      • Resources are reserved during the session even if not fully used.

• • • • Datagram data transmission is a connectionless method of sending data across a network. Each packet, called a datagram, is treated independently and may take a different path dynamically to reach the destination. There is no need to establish a dedicated path or circuit before sending data.
      • In this mode of transmission, packets from a source machine to a destination machine are transmitted and routed independently from all other packets. Here, routers decide to change the routes that different packets follow, resulting in the different packets reaching their destination in a different order from the same sender because of the different transmission routes/path lengths, differences in transmission rates, and the amount of congestion in the different paths.
      • The packets may arrive at the destination machine in an order different from the transmission order. 
      • Each data packet contains the full destination address.
      • This transmission is the foundation of the Internet and is ideal for applications where speed and flexibility are more important than guaranteed delivery or order. The Internet is a datagram network.
      • It does not create a circuit before submission.
      • Here, Network resources are not reserved for transmission.
      • Examples –
        • Early Ethernet networks/POTS (Plane Old Telephone Systems), IP, UDP, etc.
      • Advantages :

Some of the key advantages of Connectionless switching are:

• • • • • Flexibility: Here, each packet is treated independently and can take different paths through the network to reach its destination. This flexibility of transmission enables efficient utilization of network resources and better adaptability to changing network conditions.
        • Simple and Fast: It also does not need to establish or terminate connections. Hence, this transmission is simple and fast.
        • Scalability: Connectionless packet switching is highly scalable, especially in large networks with numerous nodes and high traffic. Since packets are treated independently, there is no need to establish and maintain dedicated connections between each pair of communicating nodes. This feature makes connectionless packet switching well-suited for handling a large number of simultaneous connections.
        • Fault tolerance/Robust: In connectionless packet switching, if a particular network path or node fails/congested, packets can be rerouted dynamically through alternative paths. This fault tolerance feature ensures that data transmission can continue even in the presence of network failures, resulting in improved network reliability.
        • Lower overhead: Connectionless packet switching has lower overhead compared to connection-oriented packet switching. In connection-oriented networks, resources need to be allocated and reserved upfront for the duration of the connection, which can lead to increased overhead. In contrast, connectionless packet switching does not require such upfront resource reservation, resulting in more efficient utilization of network resources.
        • Simple implementation: The absence of connection establishment and termination processes in connectionless packet switching simplifies the network implementation and reduces the complexity of network protocols. This simplicity often leads to lower costs and easier maintenance.
        • Compatibility with variable data rates: Connectionless packet switching is well-suited for networks where the data rates vary significantly among different flows or connections. It can handle different types of traffic, such as real-time multimedia streams or bursty data transfers, without the need for explicit reservation or renegotiation of connection parameters.
        • Efficient use of network: No resources are reserved, so the network can adapt to changing traffic.

• • • • Disadvantages :
        • Extra processing power is required at the nodes for attaching source and destination addresses to every datagram packet before submission, which also increases the required time of transmission.
        • Packets may arrive at the destination randomly/out of order. This requires that all the arriving packets be stored and rearranged/assembled at the buffer before final submission to the destination. For reassembly, switching requires overhead bits for indexing/numbering the unordered packets.
        • Some packets may be lost during transmission.