A LAN Switch

Download Report

Transcript A LAN Switch

Datornätverk A – lektion 11
Kapitel 16: Connecting LAN:s, Backbone
Networks and Virtual Lans.
(Kapitel 18: Frame Relay and ATM översiktligt)
Chapter 16
Connecting LANs,
Backbone Networks,
and Virtual LANs
Limitations of Ethernet
Technologies
• Distance (the length of the cable)
○ 200 m in Thin Ethernet (10Base2)
○ 100 m in twisted pair Ethernet (10BaseT or 100BaseT or Fast
Ethernet)
• Number of collisions when too many stations are
connected to the same segment
• The situation is similar in other LAN technologies
Figure 16.2
Repeater
Note:
A repeater connects segments of a
LAN.
Note:
A repeater forwards every frame bitby-bit; it has no packet queues, no
filtering capability and no collision
detection.
Figure 16.3
Function of a repeater
A repeater is a regenerator
Hubs
A hub is a multiport repeater used in
10BaseT and Fast Ethernet
Hubs give a possibility to have a
physical star topology but logical
bus topology.
Hub’s Limitations
• Hubs and repeaters resolve the problem with the distance, but does not resolve
the problem with collisions.
• A hub network can have lower throughput than several separate networks.
 The maximum
througput of the three
separate networks =
3x10Mbps
 The throughput of
the connected
network = 10Mbps
Bridges – A Simple Example
H1 H2
H3
 The frame from H1 to H4 is
forwarded by the bridge
 The frame from H1 to H3 is
dropped by the bridge
H4
LAN1
P1
B1
H5
H6
P2
LAN2
Traffic within the same group
Traffic between the two groups
Note:
A bridge has a table used in filtering
decisions.
Figure 16.5
Bridge
Figure 16.6
Learning bridge
Figure 16.7
Loop problem
Cycles in Bridged Network
1. host writes frame F
2. B1 and B2
forward the
frame, F1 and F2
are generated
to destination which is
unknown for B1 and B2
F
B1
B2
B1
F1
4. B1 and B2
forward the
frames F1 and F2
F2
F1
B1
B2
3. B2 receives F1,
B1 receives F2
B1
B2
B2
F2
F2
5. The situation in 3.
is repeated and the
frames are sent back
F1
F1
6. The frames can
circulate in the
network for ever
F2
B1
B2
B1
F1
B2
F2
Figure 16.10
Forwarding ports and blocking ports
Dotted lines = blocking (non-active redundant) ports.
May be used if one of the other bridges or links fails.
Continuous black lines = forwarding (active) ports.
These constitute a spanning tree (ett spännande träd)
without loops.
Spanning Tree Algorithm – Definitions
• Root Path Cost: For each bridge, the cost of the min-cost path to the
root.
Costs are assigned to each port or hop count is used, based on for example
bandwith, delay or number of hops (1 per port).
• Each bridge is assigned a unique identifier: Bridge ID
○ If not assigned, the lowest MAC addresses of all ports is used as the bridge
ID.
○ Low ID number means high priority.
• Each port within a bridge has a unique identifier (port ID). Typically the
MAC address of the port is used.
The Spanning Tree Algorithm
1. Elect the root bridge. (The bridge with lowest ID.)
2. Choose a root port for every bridge. (For lowest cost to
the root bridge.)
3. Chose one designated bridge for each LAN, for
minimum cost between the LAN and the root bridge.
Mark the corresponding port as a designated port.
○
○
If two bridges have the same cost, select the one with lowest ID.
If the min-cost bridge has two or more ports on the LAN, select
the port with the lowest identifier
4. Mark the root ports and designated ports as forwarding
(active) ports, the others as blocking (non-active) ports.
Figure 16.9 Applying spanning tree
Root ports: Minimum one star.
Designated ports: Two stars.
The other ports are blocking ports.
Spanning Tree - Example
1
The corresponding
graph
The network
B1
Network 1
B1
Network 2
1
Network 4
3
Network 3
B2
•
•
•
2
B2
Networks are graph nodes, ports are graph edges
A spanning tree is a connected graph which has no loops (cycles)
The dotted links are the blocked ports on the bridge, in order to prevent
loops and duplicated frames
4
Another example
B8
B3
B5
B7
B2
B1
B6
B4
Cost for each
port is 1
(hop-count)
The Root Bridge and the Spanning
Tree
**
B8
*
**
B3
**
*
Spanning Tree:
B5
**
B1
*
B7
B2
*
*
** ** **
** B1 **
Root
*
B6
**
B2
B4
B5
B7
B8
*
B4
**
**
A spanning tree is a
connected graph which
has no loops (cycles)
Multiple LANs with Bridges with
Costs Assigned
L1
4
LAN 1
B1
LAN 2
B1
Cost=2
Cost=6
B2
Cost=4
Cost=2
B3
Cost=3
Cost=4
Cost=5
B5
LAN 3
Cost=6
B4
LAN 4
Cost=5
Cost=1
L2
6
1
5
Cost=6
B6
B5
B6
2
Cost=4
4
6
L3
2 3
B3
B2 4
6
5
B4
L4
The cost of sending from L1 to
L4 via B1 and B2 is 6
Only costs for going from a
bridge to a LAN are added
Example: Root Bridge and Root Ports
L1
4
Root
4
6
B1
2
L2
B5
B6
Cost=6
5
3
2
6
Cost=6
L3
6
B3
Cost=2
B2
Cost=3
1
• Lowest cost from
each bridge to the
root bridge are
calculated.
4
5
L4
B4
Cost=8
• The root bridge
and root ports are
marked in red
Example: Designated Ports and the
Spanning Tree
*
4
Root
L1 B1
L2 2
L1
*
6
B5
B6
Cost=6
*
L2
4
Cost=3
1
5
3
2
*
L3
6
B3
6
B2
Cost=2
4 L3
B4
Cost=6
L4
*
L4
5
Cost=8
• Lowest cost from
each LAN to the
root bridge are
calculated (= the
cost from an
adjacent bridge.)
• The designated
ports are marked
“*”.
Example: Designated Ports and the
Spanning Tree
L1
4
B1
B6
2
L2
3
2
6
B2
B5
L3
B3
B4
4
L4
The rest of
the ports are
blocked.
This results in
a spanning
tree.
Figure 16.13
Connecting remote LANs
LAN Switches
• LAN switching provides dedicated,
collision-free communication between
network devices, with support for
multiple simultaneous conversations.
• LAN switches are designed to switch
data frames at high speeds.
• LAN switches can interconnect a 10Mbps and a 100-Mbps Ethernet LAN.
H1
H2
H3
H1
H3
H2
A LAN Switch
• The computer has a segment to itself
– the segment is busy only when a
frame is being transfered to or from
the computer
• As a result, as many as one-half of
the computers connected to a
switch can send data at the same
time
Figure 16.12
Star backbone
16.3 Virtual LANs
Membership
Configuration
IEEE Standard
Advantages
Figure 16.15 A switch using VLAN software
Note:
VLANs create broadcast domains.
Figure 16.16 Two switches in a backbone using VLAN software
Chapter 18
Virtual Circuit
Switching:
Frame Relay
and
ATM
Two Approaches to Packet
Switching
• Datagram networks (For example IP)
○ Analogous to the postal service
○ The inteligence is in the end devices (computers), the network should
not be trusted
○ Each packet carries the destination address
○ Destination addresses are global internationally
• Virtual circuit networks (For example X.25, Frame Relay and
ATM)
○ Analogous to the telephone service
○ The network should take all the responsibility, the end devices should
be as simple as posible
○ The path that the packets follow is determined at the beginning of the
transmission, but store and forward switching is used.
Characteristics of WANs
Circuit
Dedicated path
Continuous data
transmission
No data storage
Connection
established for
entire conversation
Call setup delay;
low transmission
delay
Busy signal
Datagram
No dedicated path
Packets
Virtual Circuit
Store and forward
Route established
for every packet
Store and forward
Route established
for every packet
No dedicated path
Packets
Packet transmission Call setup delay;
delay
Packet transmission
delay
Possible notification Notification of
of no/bad deliveries connection denial
Blocking at networkDelay at network
Blocking/delay at
overload
overload
network overload
Fixed bandwidth Dynamic bandwidth Dynamic bandwidth
No overhead/data Overhead/packet
Overhead/packet
Figure 18.1 Virtual circuit wide area network
Figure 18.3 VCI phases
Virtual Circuit Network
• Three Phases
○ Setup phase
• Network protocol establishes a logical path called virtual circuit
(VC). The path remains the same during transmission (all packets
use it)
○ Data transfer phase
• Each packet carries “tag” or “label” (virtual circuit id, VCI),
which determines next hop (the link to which the packet should
be forwarded).
• At each node, the forwarding is done by inspecting the input line,
the VCI and consulting the forwarding table at the switches.
○ Teardown phase
• All switches remove the entries about the VCI from their tables
Figure 18.2 VCI
Figure 18.4
Switch and table
X.25 Networks
• Developed in 1970s in European countries under the auspices of ITU
○ Public packet-switched networks
○ Uses virtual circuit connections
• Switched virtual circuits – analog to dial-up in circuit switching
• Permanent virtual circuits – analog to leased lines in circuit
switching.
○
○
○
○
Operates on the three lowest layers (physical, data-link and network layer)
Performs error-contol and flow-control on the node-to-node basis
Work at speed up to 64Kbps
Nowadays it is obsolete
Frame Relay
• X.25 data rates were not stisfactory for users looking for higher data
rates and lower costs
○ Checking frames for error at every node is inefficient
○ Only one fourth of traffic is message traffic, the rest is overhead
(necessary for transmission media that are more error prone)
• Frame relay – public data network that have improved performance
○ Developed having in mind new transmission media that have much lower
probability of error
○ Does not provide error checking and acknowledgement at both, the datalink layer and the network layer
X.25 versus Frame Relay
Data
Frame ack
Ack
Data
Data
Frame ack
Frame ack
switch
Ack
Data
switch
Ack
Frame ack
switch
Ack
X.25 traffic (ACKs at both data-link and transport layer)
Data
Data
Data
Data
Frame relay traffic (ACKs are required at the transport layer only)
Frame Relay in the Internet
• The virtual circuits in frame-relay are called DLCI (Data Link
Connection Identifier)
Figure 18.8
Frame Relay network
Note:
Frame Relay operates only at the
physical and data link layers.
Note:
Frame Relay does not provide flow or
error control; they must be provided by
the upper-layer protocols.
ATM – Basic Idea
• Uses small fixed-size packets called cells
○ The cells are 53 bytes long (48 bytes payload + 5 bytes header)
○ The length of the cell compromise between American and European
telephone companies (average of 32 and 64)
• Uses packet switching
○ Connection oriented (uses virtual circuits)
• Speeds of 155 Mbps or 622 Mbps are achieved over SONET
• Was heavily promoted by telephone companies as BISDN
(Broadband Integrated Services Digital Network) technology.
Figure 18.13
Multiplexing using different frame sizes
Figure 18.14
Multiplexing using cells
Note:
A cell network uses the cell as the
basic unit of data exchange. A cell is
defined as a small, fixed-sized
block of information.
ATM Basic Concepts
• Nagotiated Service Contract
○ Logical connections called Virtual circuits
• The sender nagotiates a ”requested path” with the network for a
connection to the destination
○ End-to-end Quality of Service
• When setting up a connection the sender specifies the atributes of
the call (type, sped, ...) which determine end-to-end quality of
service
• Virtual Circuit Network
○ Well defined connection procedures
○ Dedicated capacity per connection
○ Flexible access speeds
• Cell based (short packets with fixed size)
• All kinds of data look same to the network
ATM Switching
•
•


When a site has an information to send to another, it requests a connection by
sending a message
The message passes through vasious switches, setting up a virtual path
Subsequent data cells contain a virtual
path ID which the switch uses to to
Connect to B
route the cell through outgoing links
Using the input port and VP ID, the
OK
switch locates the table entry, changes End System A
the cell VP ID with one paired with
the asssociated output port and sends
the cell through that port
Connect to B
OK
OK
Connect to B
Connect to B
OK
End System B
Virtual Circuit and Paths
Virtual Circuits (VC)
ATM Physical Link
(STM-1, OC-12, E1)
Virtual Channel Connection
(VCC)
VCC - contains multiple VPs
Virtual Path (VP)
Virtual Path (VP)
Virtual Circuit (VC)
= Logical Path between
ATM End Points
VP - contains multiple VC
Figure 18.18
Example of VPs and VCs
Note:
Note that a virtual connection is
defined by a pair of numbers:
the VPI and the VCI.
Figure 18.19
Connection identifiers
Figure 18.20 Virtual connection identifiers in UNIs and NNIs
Figure 18.21 An ATM cell
Figure 18.22
Routing with a switch
ATM Service Models
• CBR (Constant Bit Rate)
○ Carries real time (constant bit rate) traffic
○ Guaranties rate, delay and loss of cells
• UBR (Unspecified Bit Rate)
○ No other guarantee besides in-order delivery of cells
• ABR (Available Bit Rate)
○ No guarantee on transmision rate, but if possible the user can use a higher
rate than in UBR.
○ Congestion feedback from the network
• VBR
○ The variable bit-rate is requested by the sender
○ Targeted toward real-time services like CBR