Switching

• Circuit-Switched Networks
• Datagram Networks
• Virtual-Circuit Networks
• Structure of a Switch

8-1
 Switching
• A switched network consists of a series of interlinked nodes, called switches.

• Switches are devices capable of creating temporary connections between two or more
devices linked to the switch.

• In a switched network, some of these nodes are connected to the end systems, others are
used only for routing.

8-2
Switched Network
 Taxonomy of Switched Networks

• In message switching, each switch stores the whole message and forwards it to the next
switch.

8-3
 Circuit-switched Networks(1)
• A circuit-switched network consists of a set of switches connected by physical links.

• A connection between two stations is a dedicated path made of one or more links.

• Each connection uses only one dedicated channel on each link.

• Each link is normally divided into n channels by using FDM or TDM.

8-4
A circuit-switched network
 Circuit-switched Networks(2)
Three phases
•Actual communication requires three phases:
- connection setup
- data transfer
- connection teardown

Setup phase
•Before the parties can communicate, a channel (circuit) is reserved on each link.

•Combination of channels defines the dedicated path.

Data transfer phase

•After the establishment of dedicated circuit (channels), the two parties can transfer data.

Teardown phase
•When one of the parties needs to disconnect, a signal is sent to each switch to release the
resources.

8-5
Circuit-Switched Network: Example 1

8-6
Circuit-Switched Network: Example 2

8-7
 Circuit-switched Networks(5)
• Not as efficient as other two types of networks.

• Delay is minimal.

• Switching at the physical layer in the traditional telephone network uses the
circuit-switching approach.

8-8
Delay in circuit-switched network
 Datagram Networks(1)

• In a packet-switched network, there is no resource allocation for a packet.

• Means that there is no reserved bandwidth on the links, and there is no scheduled
processing time for each packet.

• Resources are allocated on demand.

• Allocation is done on a first-come, first served basis.

8-9
 Datagram Networks(2)
• In a datagram network, each packet is treated independently of all others.
•
• Packets in this approach are referred to as datagrams.

• Switches in a datagram network are referred to as routers.

• Datagrams may arrive at their destination out of order.

• Datagrams may also be lost or dropped because of a lack of resources.

• Sometimes referred to as connectionless networks.

• Switching in the Internet is done by using datagram approach at the network

layer.

8-10
 Datagram Networks(3)

Routing Table
• A switch in a datagram networks uses a routing table
that is based on the destination address.

• Routing tables are dynamic and are updated

periodically.

• Every datagram carries a header.

• The destination address in the header of a datagram

remains the same during the entire journey of the
packet 8-11
 Datagram Networks(4)
• Efficiency of a datagram network is better than of a circuit-switched network.

• Datagram network may have greater delay than a virtual-circuit network even
though no setup and teardown phases.

• Each datagram may experience a wait at a switch before it is forwarded.

• Delay is not uniform.

8-12
Delay in datagram network
 Virtual-Circuit Networks(1)
• A virtual-circuit network (VCN) is a cross between a circuit-switched network
and a datagram network.

• It has some characteristics of both:

- there are setup, data transfer, and teardown phases as in a circuit-
switched network (CSN)
- resources can be allocated during setup phase, as in a CSN, or on
demand as in a datagram network (DN)
- As in DN, data are packetized and each packet carries an address in
the header. This address is local i.e., it is the address of the next switch
not the end system
- As in CSN, all packets follow the same path established during the
connection
- VCN is normally implemented in the data link layer, while CSN is in
physical layer and DN in the network layer
8-13
Virtual-Circuit Network

8-14
 Virtual-Circuit Networks(3)
Addressing
•Two types of addressing: global and local (virtual-circuit identifier: VCI)

•Global address is used only to create a VCI.

•Virtual Circuit Identifier

- the identifier that is actually used for data transfer is called the VCI.
- unlike global address, it is a small number, used by a frame between two
switches.

8-15
 Virtual-Circuit Networks(4)
Three Phases
•Data transfer phase, setup phase, teardown phase

Data Transfer Phase

•To transfer a frame from S to its D, all switches need to have a table entry for this
virtual-circuit.

Setup Phase
•Switch creates an entry for a virtual-circuit.
•Two steps are required: the setup request and the acknowledgement.

Teardown Phase
•Two frames are exchanged: the teardown request and teardown confirmation.

8-16
 Data transfer Phase

• Source-to-destination data transfer in a virtual-circuit network

8-17
Setup Phase: Setup Request

8-18
Setup Phase: Acknowledgement

8-19
 Virtual-Circuit Networks(8)
• In virtual-circuit switching, all packets belonging to the same source and destination
travel the same path; but the packets may arrive at the destination with different
delays if resource allocation is on demand.

• Switching at the data link layer in a switched WAN is normally implemented by using
virtual-circuit techniques.

8-20
 Structure of Circuit Switches
• Space-division switch or time-division switch.

Space-division Switch
- in this switching, the paths in the circuit are separated from one another
spatially.
- crossbar switch, multistage switch

8-21
Crossbar Switch
 Crossbar Switches: Problem
• The number of switches is huge.
– connect n inputs by m output -- requires n*m crosspoints.
– Ex : 1000 input, 1000 output → 1,000,000 crosspoint

• Inefficient
– fewer than 25% of the crosspoints are in use at a given
time.

8-22
 Multistage Switch

• Solution to the limitation of the crossbar switch.

• Combines crossbar switches in several stages (normally 3).

8-23
 Design of a Three-Stage Switch
• Three steps
– Divide the N input lines into groups, each of n lines. For each group, use one
crossbar of size n x k, where k is the number of crossbars in the middle stage
– Use k crossbars, each of size (N/n) x (N/n) in the middle stage
– Use N/n crossbars, each of size k x n at the third stage.

• N/n(n x k) + k(N/n x N/n) + N/n(k x n) = 2kN + k(N/n)2

= Total no. of crosspoints

• In a three-stage switch, the total number of crosspoints is much

smaller than the number of crosspoints in a single-stage switch (N 2).

8-24
 Multistage Switch: Example 1
• Design a three-stage, 200 × 200 switch (N = 200) with k =
4 and n = 20.

• In the first stage we have N/n or 10 crossbars, each of size

20 × 4. In the second stage, we have 4 crossbars, each of
size 10 × 10. In the third stage, we have 10 crossbars,
each of size 4 × 20. The total number of crosspoints is 2kN
+ k(N/n)2, or 2000 crosspoints. This is 5 percent of the
number of crosspoints in a single-stage switch (200 × 200
= 40,000).

8-25
 Multistage Switches
• Advantage: The number of crosspoints.

• Disadvantage: blocking
(Blocking refers to times when one input cannot be connected
to an output because there is no path available between them –
all the possible intermediate switches are occupied.)

• Clos criterion: condition of nonblocking

– n = (N/2)1/2
– k > 2n – 1
– Crosspoints ≥ 4N [(2N)1/2 – 1]

8-26
 Multistage Switch: Example 2
• Redesign the previous three-stage, 200 × 200 switch, using the Clos
criteria with a minimum number of crosspoints.
.
• We let n = (200/2)1/2, or n = 10. We calculate k = 2n − 1 = 19. In the
first stage, we have 200/10, or 20, crossbars, each with 10 × 19
crosspoints. In the second stage, we have 19 crossbars, each with 10
× 10 crosspoints. In the third stage, we have 20 crossbars each with
19 × 10 crosspoints. The total number of crosspoints is 20(10 × 19) +
19(10 × 10) + 20(19 ×10) = 9500.

8-27
 Time-division switch
• Time Division Switching uses TDM inside a switch.

• Time-slot interchange (TSI) – most popular.

• The size of each location is same as the size of a single time slot.

• The no. of locations is the same as the no. of inputs.

• The RAM fills up with incoming data from time slots in the order received. Slots are
then sent out in an order based on the decisions of a control unit.

8-28
Time-Slot Interchange
Time- and Space-Division Switch Combinations

• Advantage of space division switching is that it is instantaneous, requires more

no. of crosspoints.

• Advantage of time division switching is that it needs more no crosspoints, but

introduces delay.

• To take advantage of best of both, combine space-division and time-division

technologies.

• Fig. shows a simple time-space-time (TST) switch that consists of two time
stages and one space stage.

• Last stage is a mirror image of the first stage.

• Middle stage is a space-division switch (crossbar) that connects the TSI groups
to allow connectivity between all possible input and output pairs.

8-29
TST Switch

8-30
 Structure of Packet Switches

• Four components: input port, output port, routing processor, and

switching fabric.

8-31
 Structure of Packet Switches(2)
INPUT PORTS
• Perform physical and data link functions.
• Bits are constructed from the received signal.
• Packet is decapsulated from the frame.
• Errors are detected and corrected.
• In addition to physical layer processor and a data link processor, the input port
has buffers (queues) to hold the packet before it is directed to the switching
fabric.

8-32
 Structure of Packet Switches(3)
OUTPUT PORTS
• Perform the same functions as the input port, but in the reverse order.
• First, outgoing packets are queued, then the packet is encapsulated in a frame,
and finally to create the signal to be sent on the line.

8-33
 Structure of Packet Switches(4)
ROUTING PROCESSOR
• Perform the functions of network layer.
• Table lookup: searching the routing table for next hop address and output
port number by destination address.

SWITCHING FABRICS
• Difficult task in a packet switch is to move the packet from the input queue to
the output queue.
• The speed with which this is done affects the size of the input/output queue
and the overall delay in packet delivery.
• Types: Crossbar switch
Banyan switch
Batcher-Banyan switch

8-34
 Banyan Switch
• Multistage switch with microswitches at each stage that route the packets
based on the output port represented as a binary string.

• For n inputs and n outputs, we have log2n stages with n/2 microswitches
at each stage.

8-35
 Examples of Routing in a Banyan Switch
• First stage routes the packet based on the high-order bit of the binary
string.

• Second stage routes the packet based on the second high-order bit, and so
on.

8-36
 Batcher-Banyan Switch
• Problem in Banyan switch: possibility of internal collision
– Solve it by sorting the arriving packets based on their destination port
• Sorting switch uses hardware merging techniques.
• Trap module prevents duplicate packets from passing to the banyan
switch simultaneously.

8-37