Download Ethernet

LAN (Ethernet), Multicast
Why Multicast
• When sending same data to multiple receivers
– better bandwidth utilization
– less host/router processing
– quicker participation
• Application
–
–
–
–
–
Video/Audio broadcast (One sender)
Video conferencing (Many senders)
Real time news distribution
Interactive gaming
Cluster computing
IP multicast service model
• Invented by Steve Deering (PhD. 1991)
– It’s a different way of routing datagrams
•
•
•
•
•
•
RFC1112 : Host Extensions for IP Multicasting - 1989
Senders transmit IP datagrams to a "host group"
“Host group” identified by a class D IP address
Members of host group could be present anywhere in the Internet
Members join and leave the group and indicate this to the routers
Senders and receivers are distinct: i.e., a sender need not be a
member
• Routers listen to all multicast addresses and use multicast routing
protocols to manage groups
Class D: Multicast IP addresses
IP Multicast Addresses
IGMP Snooping
• Internet Group Management Protocol (IGMP RFC 2236) used to manage IP multicast traffic
• Application wishing to receive traffic for specific
IP multicast address sends out an ICMP join
request (or a leave request to stop receiving
multicast)
• Switches that employ IGMP snooping listen for
IGMP join/leave requests to decide when to send
a specific multicast frame to a port
IGMP – Joining a group
IGMP Membership-Report
R
Network A
Example : R joins to Group 224.2.0.1
•
R sends IGMP Membership-Report
about 224.2.0.1 to 224.0.0.2
(ALL-ROUTERS.MCAST.NET)
•
DR receives it. DR will start
forwarding packets for 224.2.0.1 to
Network A
•
DR periodically sends IGMP
Membership-Query to 224.0.0.1
(ALL-SYSTEMS.MCAST.NET)
•
R answers IGMP MembershipReport about 224.2.0.1
DR
Network B
Data to 224.2.0.1
R: Receiver
DR: Designated Router
IGMP – Leaving a group
IGMP Leave-Group
Example : R leaves from a Group 224.2.0.1
R
•
R sends IGMP Leave-Group to
224.0.0.2
(ALL-ROUTERS.MCAST.NET)
•
DR receives it.
•
DR stops forwarding packets for
224.2.0.1 to Network A if no more
224.2.0.1 group members on Network
A.
Network A
DR
Network B
Data to 224.2.0.1
R: Receiver
DR: Designated Router
Protocol Independent Multicast
• PIM : Protocol Independent Multicast
– Independent of particular unicast routing protocol
– Most popular multicast routing protocol today
• PIM supports both dense (DM) and sparse (SM)
mode operation
– Opt out (NACK) type (DM)
• Start with “broadcasting” then prune brunches with no receivers, to
create a distribution tree
• Lots of wasted traffic when there are only a few receivers and they are
spread over wide area
– Opt in (ACK) type (SM)
• Forward only to the hosts which explicitly joined to the group
• Latency of join propagation
PIM DM overview
• Assumes that you have lots of folks who want to
be part of a group
• Based on broadcast and prune
– Ideal for dense group
•
•
•
•
Source tree created on demand based on RPF rule
If the source goes inactive, the tree is torn down
Easy “plug-and-play” configuration
Branches that don’t want data are pruned
PIM DM overview
• Grafts used to join existing source tree
• Asserts used to determine the forwarder for
multi-access LAN
• Non-RPF point-2-point links are pruned as
a consequence of initial flooding
PIM-DM(1)
Initial flood of data
S
A
Source
B
G
C
F
D
H
E
I
R1
Receiver 1
R2
Receiver 2
RPF(reverse path forwarding)
•
Simple algorithm developed to avoid duplicate packets
on multi-access links
•
RPF algorithm takes advantage of the IP routing table to
compute a multicast tree for each source.
•
RPF check
1. When a multicast packet is received, note its source (S) and
interface (I)
2. If I belongs to the shortest path from S, forward to all
interfaces except I
3. If test in step 2 is false, drop the packet
•
Packet is never forwarded back out the RPF interface!
PIM-DM(2)
prune non-RPF p2p link
S
A
IGMP PIM-Prune
Source
B
G
C
F
D
H
E
I
R1
Receiver 1
R2
Receiver 2
PIM-DM(3)
C and D Assert to Determine
Forwarder for the LAN, C Wins
S
A
IGMP PIM-Assert
with its own IP address
Source
B
G
C
F
D
H
E
I
R1
Receiver 1
R2
Receiver 2
PIM-DM(4)
I, E, G send Prune
H send Join to override G’s Prune
S
A
IGMP PIM-Prune
IGMP PIM-Join
Source
B
G
C
F
D
H
E
I
R1
Receiver 1
R2
Receiver 2
PIM-DM(5)
I Gets Pruned
E’s Prune is Ignored (since R1 is a receiver)
G’s Prune is Overridden (due to new receiver R2)
S
A
Source
B
G
C
F
D
H
E
I
R1
Receiver 1
R2
Receiver 2
PIM-DM(6)
New Receiver, I send Graft
S
A
IGMP PIM-Graft
Source
B
G
C
F
D
H
I
E
R1
R2
Receiver 1
R3
Receiver 3
Receiver 2
PIM-DM(6)
new branch
S
A
IGMP PIM-Graft
Source
B
G
C
F
D
H
I
E
R1
R2
Receiver 1
R3
Receiver 3
Receiver 2
Multicast Scope Control:
TTL Boundaries
to keep multicast traffic within an administrative
domain, e.g., for privacy or resource reasons
the rest of the Internet
an administrative domain
TTL threshold set on
interfaces to these links,
greater than the diameter
of the admin. domain
Direct connection: broadcast
• Shared media
Metcalfe’s Ethernet
Sketch (1973)
Ethernet “dominant” LAN technology:
•
•
•
•
•
cheap $30 for 100Mbs!
first widely used LAN technology
simpler, cheaper than token LANs and ATM
kept up with speed race: 10, 100, 1000, 10000 Mbps
wireless options
10Mb/s Ethernet Physical Layer
• Each bit has a transition
• Allows clocks in sending and receiving nodes to
synchronize to each other
– no need for a centralized, global clock among nodes!
Ethernet Format: Framing
• Preamble: (synchronization)
– 8 bytes, allows sender/receiver clocks to synchronize
•
Destination/Source Address: (hey Paul, Tom here)
– 6 bytes each
• Type:
– 2 bytes, indicates higher layer protocol
– 0x0800 is IP, 0x0806 is ARP
• Data: 46-1500 bytes
• FCS (CRC):
– catches most transmission errors - errored frames dropped
Ethernet Packet Structure
•14 byte header
•2 addresses
Ethernet Addressing
• 6 byte address (unique to each adapter)
– Example: 08-0b-db-e4-b1-02
– 2^48 = 281 trillion; can produce 100 million LAN devices every
day for 2000 years!
• Interpretation of address:
–
–
–
–
–
Upper 24 bits OUI (Organizationally Unique Identifier)
Lower 24 bits Organization-assigned portion
Unicast: lowest bit of first byte is 0
Multicast: lowest bit of first byte is 1
Broadcast: ff-ff-ff-ff-ff-ff
• Adaptor accept frame if and only if:
– Destination address matches adapter address, or
– Destination address is broadcast, or
– Destination address is multicast and adapter has been configured
to accept it
Ethernet Media sharing
• CSMA/CD (the polite
conversationalist)
– carrier sense: don’t transmit if
you sense someone else
transmitting
– collision detection: abort your
transmission if you sense
someone else transmitting
– random access: wait random
time before attempting a
retransmission
Ethernet Technologies
• 10Base2:
– 10Mbps, 200 meters max cable length
– thin coaxial cable in a bus topology
– repeaters connect multiple segments
• 10BaseT / 100BaseT “fast ethernet”:
– 10/100Mbps, Twisted pair
– Nodes connect to a hub in “star topology”
nodes
• Gigabit Ethernet:
– 1Gbps, fibre or copper
– Extending from LAN to MAN
• 10 Gbps Ethernet available
• High data speed + larger distance + increasing
number of devices per LAN => switching
hub
Twisted Pair Wire Map
• EIA/TIA 568B (UGA Standard)
Standard vs Crossover Cables
Card-to-Hub Wiring
(Standard Cable)
TD+
TDRD+
RD+
RDTD+
RD-
TD-
Card-to-Card (Hub-to-Hub) Wiring
(Crossover Cable)
TD+ (RD+)
TD- (RD-)
RD+ (TD+)
TD+ (RD+)
TD- (RD-)
RD+ (TD+)
RD- (TD-)
RD- (TD-)
Power over Ethernet (PoE)
http://www.nwfusion.com/news/2003/1124infrapoe.html
Ethernet
IP: 10.0.0.10
IP: 10.0.0.11
MAC: 00:00:aa:aa:aa:aa
MAC: 00:00:bb:bb:bb:bb
A
B
C
D
IP: 10.0.0.12
IP: 10.0.0.13
MAC: 00:00:cc:cc:cc:cc
MAC: 00:00:dd:dd:dd:dd
• Most popular LAN technology
nowadays 10Mb/s - 1Gb/s
• Each host has unique 48bit
MAC address (factory assigned)
• Frames sent to MAC addresses
• To find destination MAC
address, ARP protocol is used
Ethernet frame
Dest
MAC
Source
MAC
Dest
IP
Source
IP
IP packet
Data
ARP: finding the MAC Address
Host A
ARP Query
Broadcast
Host B
MAC ?
Host B
Host B
IP
ARP Response
Host B
Unicast
MAC
Host B
IP
RFC 826: Address Resolution Protocol, 1982
ARP frame format
Multicast: one to many communication
• Application level one to
many communication
• multiple unicasts
• IP multicast
R
S
R
S
R
R
R
R
IP & Ethernet Multicast Address
Mapping
• IP multicast addresses (class D) range from
224.0.0.1 to 239.255.255.255 and map to
Ethernet destination MAC addresses as
shown below
32-bit Class D IP Address
1110
Low-order 23 bits of multicast
Not mapped
Group ID copied to Enet address
00000001 00000000 01011110 0
48-bit Ethernet Address
Multicast Addresses
• Multicast revises addresses to be protocol
specific: high byte, least bit is “1” if multicast.
Multicast(1)
high
byte
Local(1)/global(0)
administration
48 bit address
• Applications that use multicast
– One-to-many IP video broadcasting
– Computing clusters in Grids
Ethernet Multicast Addresses
01-00-5E-00-00-00
Switching (same as Bridging)
• Goals
– traffic isolation
– “transparent” operation
– plug-and-play
• Operation
– store and forward Ethernet frames
– examine frame header and selectively forward frame based
on MAC dest address
– when frame is to be forwarded on segment, uses CSMA/CD
to access segment
Switching Tables
0260.8c01.1111
E0:
E0:
E1:
E1:
E0
0260.8c01.2222
0260.8c01.1111
0260.8c01.2222
0260.8c01.3333
0260.8c01.4444
0260.8c01.3333
E1
0260.8c01.4444
Spanning Tree Protocol
X
Y
Segment 1
Broadcast
Segment 2
Spanning tree protocol (IEEE 802.1d)
• Every bridge has bridge-id
– bridge-id = 2-byte priority + 6-byte MAC addr
• MAC address is 00:A0:C5:12:34:56
• bridge ID is 8000:00A0:C512:3456
• Every port of bridge has
– port-id = 1-byte priority + 1-byte port-number
– port-cost = inversely proportional to link speed
• Bridge with lowest bridge-id is root bridge
• On each LAN segment, bridge with lowest path cost to
root is designated bridge (use bridge-id and port-id to
break ties)
• A bridge forwards frames through a port only if it is a
designated bridge for that LAN segment
Spanning tree example
A
B
B3
DP
DP
C
B5
DP
B2
E
RP
D
B7
RP
RP
DP
DP
G
DP
B1
root
DP
F
DP
DP
H
RP
B6
I
B4 DP
DP
J
K
STP terminology
• Port roles:
– Root port (switch port leading to root)
– Designated port (LAN port leading to root)
– Alternate / backup port (anything else)
• Port states:
–
–
–
–
Blocking (no send/rcv, except STP bpdus)
Listening (prepare for learning/forwarding)
Learning (learn MAC addr but no forwarding)
Forwarding (send/rcv frames)
• Can disable STP on port or switch
– All frames are forwarded
– BPDUs?
New Spanning Tree Protocol versions
 Implementation of :
•Rapid Spanning Tree Protocol 802.1w (RSTP);
•Per VLAN Spanning Tree 802.1q (PVST +);
•Multiple Spanning Tree 802.1s (MST);
•Load balancing across links;
•Uni-Directional Link Detection (UDLD)
Virtual LANs
• LAN (broadcast domain) grows large
• “departments” or “workgroups” not happy with
big broadcast domain
– Security (eavesdropping)
– Bandwidth consumed by flooding/multicasting
• Split LAN into multiple broadcast domains
– Multiple physical LANs?
• Too expensive!
• People move all the time!
• VLAN: logical partition of LAN
Virtual LANs
VLANs: IEEE 802.1q
destination
addr
source
addr
VLAN protocol id
= 0x8100
type
data
FCS
3-bit priority
1-bit CFI
12-bit VLAN id
• “Tagged” Ethernet frames contain VLAN-id
• Switch adds/removes tag when forwarding frames between
trunk and non-trunk ports
• Complications:
– Hosts and legacy switches do not understand VLAN tags
– Tag insertion/removal requires FCS recomputation
– Frame length increases beyond legacy MTU
VLAN Standard: IEEE 802.1q
CFI-Canonical Format Identifier (Ethernet/TokenRing)
The 802.3 (legacy) and 802.1Q Ethernet
frame formats
L2 Tunneling
The default system MTU for traffic on the switch is 1500 bytes. You can configure the switch to support
larger frames by using the system mtu global configuration command. Because the 802.1Q tunneling
feature increases the frame size by 4 bytes when the metro tag is added, you must configure all switches in
the service-provider network to be able to process larger frames by increasing the switch system MTU size
to at least 1504 bytes. The maximum allowable system MTU for Catalyst 3550 Gigabit Ethernet switches is
2000 bytes; the maximum system MTU for Fast Ethernet switches is 1546 bytes.
Some Switches Support Priorities
802.1p Prioritization
• Eight levels of prioritization - p0 (lowest)
through p7 (highest)
• 802.1p example
VLAN/802.1p Switch
FS
Internal Queues:
FS
p7: VS
VS
p0: FS
FS
VS
VS
VS
VS
VS
VS
L2 Switch
Gigabit Ethernet over Fiber
Switch Configuration Example
interface GigabitEthernet2/9
description NISN/NASA
mtu 9216
no ip address
speed nonegotiate
switchport
switchport trunk encapsulation dot1q
switchport trunk allowed vlan 210-213,217-226,231,232
switchport mode trunk
switchport nonegotiate
interface FastEthernet0/7
description ASSA
switchport access vlan 210
no ip address
speed 10