Transcript ATM

NETE0510
ATM
Dr. Supakorn Kungpisdan
[email protected]
NETE0510: Communication Media and Data
Communications
1
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
2
ATM Overview
 Asynchronous Transfer Mode
 ATM is a competing technology with Ethernet switching,
but the areas of application for these two technologies
only partially overlap
 ATM is a connection-oriented, packet-switched
technology
 Use virtual circuits
 Connection phase is called “signaling”
 The main signaling protocol is known as Q.2931
 Q.2931
 Discover a suitable route across an ATM network
 Responsible for allocating resources at the switches among the
circuit
 This is to ensure the circuit a particular quality of service (QoS),
one of the greatest ATM strengths
NETE0510: Communication Media and Data
Communications
3
ATM Overview (cont’d)
 ATM has specific format of identifying addresses: E.164
and NSAP (network service access point), different from
MAC address in LANs
 Packets that are switched in ATM has fixed length: 53
bytes
 5 bytes of header followed by 48 bytes of payload
 To distinguish between fixed-length packets and
variable-length packets, they are given a special name
“cells”
NETE0510: Communication Media and Data
Communications
4
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
5
Cells
 Variable-length packets are normally constrained
to fall within some bounds
The lower bound is set by the minimum amount of
information that needs to be contained in the packet
 Typically a header with no optional extensions
The upper bound may be set by a variety of factors
 Maximum FDDI packet size
• How long each station is allowed to transmit without passing on
the token, and thus determine how long a station might have to
wait for the token to reach it
 Cell are both fixed-length and small in size
NETE0510: Communication Media and Data
Communications
6
Cell Size
 Cell has fixed length in order to facilitate the
implementation of hardware switches
 When ATM was being created in the mid- and
late 1980s, 10-Mbps was the cutting-edge
technology in terms of speed.
 To go much faster, most people thought in terms
of hardware
 Fixed-length packets turn out to be very helpful
thing if want to build fast, highly scalable
switches
NETE0510: Communication Media and Data
Communications
7
Cell Size (cont’d)
Two main reasons:
Easier to build hardwa5re to do simple jobs,
and the job of processing packets is simple
when you already know how long each one will
be
If all packets are the same length, then you can
have lots of switching elements all doing much
the same thing in parallel, each of them taking
the same time to do its job
NETE0510: Communication Media and Data
Communications
8
Cell Size (cont’d)
 The second reason improves scalability of
switch designs
Ease the task of building such hardware and that there
was a lot of knowledge available about how to build cell
switches in hardware at the time the ATM standards
were being defined
 Fixed-length cell is also good for queuing
Queues build up in a switch when traffic from several
inputs may be heading for a single output
Once you extract a packet from a queue and start
transmitting it, you need to continue until the whole
packet is transmitted
NETE0510: Communication Media and Data
Communications
9
Cell Size (cont’d)
 The longest time that a queue output can be tied up is
equal to the time it takes to transmit a maximum-sized
packet
 Tqueue = Ttrans-max-size-packet
 Fixed-length cell means that a queue output is never tied
up for more than the time it takes to transmit one cell,
which is almost certainly shorter than the maximumsized packet on a variable-length packet network
 Tqueue-cell ≤ Ttrans-max-size-packet
 If tight control over the latency that is being experienced
by cells is important, cells provide some advantage
NETE0510: Communication Media and Data
Communications
10
Cell Size (cont’d)
 A network with variable-length packets,
 Maximum packet size is 4kB
 Link speed is 100 Mbps
 Thus the time to transmit a max-size packet is 4096 x 8 /100M =
327.68 µs
 A high-priority packet that arrives just after the switch starts to
transmit a 4-kB packet will have to wait in a queue 327.68 µs
 In contrast, if the switch were forwarding 53-byte cells, the longest
wait would be 53 x 8 /100M = 4.24 µs
 Also queues of cells also tend to be a little shorter than queues of
packets
 When a packet begins to arrive in an empty queue, switch has to wait
for the whole packet to arrive before it can start transmitting the packet
on an outgoing link
 The link sits idle when the packet arrives
 Large-size packet VS small-size cell
NETE0510: Communication Media and Data
Communications
11
How large is a cell size?
 If the size is too short, the amount of header relative to the amount
of data gets larger
 Percentage of link bandwidth used to carry data goes down
 If you build up a device that processes cells at some max number of
cells per second, then as cells get shorter, the total data rate drops
in direct proportion to cell size
 E.g. network adapter that reassembles cells into larger units before
handling them up to the host
 The performance of such device depends on cell size
 If the cell is too big, there is a problem of waste bandwidth caused
by the need to pad transmitted data to fill a complete cell
 If the payload size is 48 bytes and you want to transmit 1 byte of data 
need to pad 47 bytes
 This lowers link utilization
 53-byte size were chosen, 5 bits of header and 48 bits of payload
NETE0510: Communication Media and Data
Communications
12
Cell Format
 Two different cell formats
 UNI (user-network interface): between a telephone company and
its user
 NNI (network-network interface): between a pair of telephone
companies
 The only significant difference in cell formats is that the NNI format
replaces the GFC fielded with 4 extra bits of VPI
 Understanding all the three-letter acronyms (TLAs) is a key part of
understanding ATM
ATM Cell Format at the UNI
NETE0510: Communication Media and Data
Communications
13
ATM Cell Format (cont’d)
 GFC (Generic Flow Control)
4 bits, not widely used, were intended to have local
significance at a site and could be overwritten in the
network and could be overwritten in the network
To provide a means to arbitrate access to the link if the
local site used some shared medium to connect to ATM
 VPI (Virtual Path Identifier) and VCI (Virtual
Circuit Identifier)
8-bit VPI and 16-bit VCI
Both are used to identify a virtual connection
NETE0510: Communication Media and Data
Communications
14
ATM Cell Format (cont’d)
 Type
3 bits  8 possible values
4 of them, where the 1st bit is set, relate to
management functions
If the 1st bit is not set, the cell contains user data
The 2nd bit is the “explicit forward congestion
indication” (EFCI) bit
 Used for congestion control in conjunction with the available
bit rate (ABR) service  set by the congested switch to tell
an end node that it is congested
The 3rd bit is the “user signaling” bit
 Used in conjunction with ATM adaptation Layer 5 to delineate
frames
NETE0510: Communication Media and Data
Communications
15
ATM Cell Format (cont’d)
 CLP (cell loss priority)
 A user or network element may set this bit to indicate cells that
should be dropped preferentially in the event of overload
 E.g. a video coding application could set this bit for cells that
would not dramatically degrade the quality of the video if
dropped
 A network element might set this bit for cells that have been
transmitted by a user in excess of the amount that was
negotiated
 CRC
 Known as header error check (HEC), use CRC-8
 Provide error detection and single-bit error correction capability
on the cell header only
 Important because error in the VCI will cause the cell to be
misdelivered
NETE0510: Communication Media and Data
Communications
16
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
17
Segmentation and Reassembly
 The packet handed down from high-level protocol are
often larger than 48 bytes  need fragmentation and
reassembly
 In ATM, often called segmentation and reassembly (SAR)
 To deal with this, a protocol was added in between ATM
and the variable-length packet protocols that might use
ATP, such as IP
 This layer is called ATM Adaptation Layer (AAL)
 AAL header contains the information needed by the
destination to reassemble the individual cells back into
the original message
NETE0510: Communication Media and Data
Communications
18
AAL and ATM
Segmentation and reassembly in ATM
NETE0510: Communication Media and Data
Communications
19
AALs
 ATM supports all sorts of services, including voice,
video, and data
 Need different AALs
 4 adaptation layers was originally defined:
 1 and 2 designed to support applications, like voice, that require
guaranteed bit rates
 3 and 4 support packet data running over ATM
 AAL3 used by connection-oriented packet services (e.g. X.25)
 AAL4 used by connectionless services (e.g. IP)
 Later 3 and 4 were combined into AAL3/4
 Some perceived shortcomings in AAL3/4 caused AAL5
 Now there are 4 AALs: 1, 2, 3/4, and 5
NETE0510: Communication Media and Data
Communications
20
ATM Adapter Layer 3/4
 Main function of AAL3/4 is to provide enough information
to allow variable-length packets to be transported across
the ATM network as a series of fixed-length cells
 A packet is this layer is called “protocol data unit
(PDU)”
 The task of segmentation/reassembly involves two
different packet formats:
 Convergence sublayer PDU (CS-PDU): defines a way of
encapsulating variable-length PDUs prior to segmenting them
into cells
 The PDU passed down to the AAL is encapsulated by adding a
header and a trailer, and the resultant CS-PDU is segmented
into ATM cells
NETE0510: Communication Media and Data
Communications
21
AAL3/4 CS-PDU Packet Format
ATM AAL3/4 packet format
 CPI indicates which version of the CS-PDU format is in use
 Only the value 0 is currently defined
 Beginning tag (Btag) is supposed to match the end tag (Etag) for
a given PDU
 Protect against the situation in which the loss of the last cell of one PDU
and the first cell of another causes two PDUs to be inadvertently joined
into a single PDU and passed up to the next layer in the protocol stack
 Buffer allocation size (BASize) is not necessarily the length of the
PDU
 Supposed to be a hint to the reassembly process as to how much buffer
space to allocate for the reassembly
NETE0510: Communication Media and Data
Communications
22
AAL3/4 Packet Format (cont’d)
 Before adding the CS-PDU trailer, the user data is padded to one
byte less than a multiple of 4 bytes, by adding up to 3 bytes of
padding
 Ensure that the trailer is aligned on a 32-bit boundary, making for
more efficient processing
 The CS-PDU trailer contains the Etag and the real length of the
PDU (Len)
 AAL3/4 specifies a header and trailer that are carried in each cell
ATM cell format for AAL3/4
 The CS-PDU is segmented into 44-byte chunks plus 4 bytes of its
header and trailer  bring up to 48 bytes
 Then carried as the payload of an ATM cell
NETE0510: Communication Media and Data
Communications
23
AAL3/4 Packet Format (cont’d)
 Type
 The first two bits of the AAL3/4 header contain Type field
indicating is this is the first cell of a CS-PDU, the last cell of a
CS-PDU, a cell in the middle of a CS-PDU, or a single-cell PDU
(in which case it is both first and last)
NETE0510: Communication Media and Data
Communications
24
AAL3/4 Packet Format (cont’d)
 SEQ (sequence number)
 Detect cell loss or misordering so that reassembly can be
aborted
 MID (multiplexing identifier)
 Multiplex several PDUs onto a single connection
 Length
 Show the number of bytes of PDU that are contained in the
cell
 Must equal 44 for BOM and COM cells
 CRC
 Detect errors anywhere in the 48-byte cell payload
NETE0510: Communication Media and Data
Communications
25
Encapsulation and Segmentation for
AAL3/4
NETE0510: Communication Media and Data
Communications
26
ATM Adaptation Layer 5
 AAL3/4 seems to take a lot of fields and thus a lot of
overhead to perform conceptually simple function of SAR
 Could just have 1 bit in the ATM header (as opposed to
the AAL header) to delineate the end of a frame, them
SAR could be accomplished without using any of the 48byte ATM payload for SAR information
 This led to AAL5
NETE0510: Communication Media and Data
Communications
27
AAL5 (cont’d)
 Replace 2-bit Type filed of AAL3/4 with 1 bit of
framing information in the ATM cell header
Set this bit to identify the last cell of a PDU
The next cell is assumed to be the first cell of the next
PDU
Subsequent cells are assumed to be COM cells until
another cell is received with this bit set.
 Other features of AAL3/4 are provided in AAL5
NETE0510: Communication Media and Data
Communications
28
AAL5 CS-PDU
 AAL5 CS-PDU consists of data portion and 8-byte trailer
 To make sure that the trailer always falls at the tail end of
an ATM cell, there may be up to 47 bytes of padding
between the data and trailer
 The first 2 bytes of trailer is Reserve and must be 00
 Len is the number of bytes carried in the PDU, not
including the trailer or any padding before trailer
ATM AAL5 packet format
NETE0510: Communication Media and Data
Communications
29
Encapsulation and Segmentation for
AAL5
NETE0510: Communication Media and Data
Communications
30
Encapsulation and Segmentation for
AAL5 (cont’d)
 AAL5 provides almost the same functionality as AAL3/4
without using an extra 4 bytes out of every cell
 The main feature missing from AAL% is the ability to
provide an additional layer of multiplexing onto one
virtual circuit using the MID
 AAL5 is preferred in the IETF form transmitting IP
datagrams over ATM
 More efficient use of bandwidth and simple design are
the main features that make it more appealing than
AAL3/4
NETE0510: Communication Media and Data
Communications
31
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
32
Virtual Paths
 ATM uses 24-bit identifier for virtual circuits
 8-bit virtual path identifier (VPI)
 16-bit virtual circuit identifier (VCI)
 This creates two levels of virtual connections
 A corporation has two sites connecting to a public ATM
network and each site the corporation has a network of
ATM switches
 We can establish a virtual path between two sites using
only VPI field
 The switch in the public network would use the VPI as the
only field on which to make forwarding decisions
 It is a virtual circuit network with 8-bit circuit identifier
 The 16-bit VCI has no interest to these public switches
NETE0510: Communication Media and Data
Communications
33
Virtual Paths (cont’d)
 Within corporate sites, the full 24-bit space is used for
switching
 Traffic flowing between two sites is routed to a switch
that has an connection to the public network, and its top
(most significant bits) 8 bits VPI are mapped onto the
appropriate value to get the data to the other site
NETE0510: Communication Media and Data
Communications
34
Virtual Paths (cont’d)
 Advantages
Support many virtual connections across the public
network
The switches in the public network behave as if there is
only one connection
There needs to be much less connection-state
information stored on the switches, avoiding the need
for big, expensive tables of per-VCI information
NETE0510: Communication Media and Data
Communications
35
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
36
Physical Layers for ATM
 From early in the process of standardizing ATM, it was
assumed that ATM would run on top of a SONET
physical layer
 Standard ways of carrying ATM cells inside a SONET
frame have been defined. Can buy ATM-over-SONET
products
 However, they are entirely separable.
 Can lease a SONET link from a telephone company to carry
variable-length packets
 Can send ATM cells over many other physical layers instead of
SONET
 Notable early physical layers for ATM was TAXI, the physical
layer used in FDDI
 Wireless physical layers for ATM are also being defined
NETE0510: Communication Media and Data
Communications
37
How to find boundaries of the ATM cells
 With SONET, find ATM cell boundaries:
 One of the overhead bytes in the SONET frame can
be used as a pointer into the SONET payload to start
of an ATM cell
 Having found the start of one cell, it is known that the
next cell starts 53 bytes further on in the SONET
payload
 In theory, need to read this pointer only once, but in
practice, may read it every time the SONET overhead
goes by to detect errors or resynchronize if needed
NETE0510: Communication Media and Data
Communications
38
Outline
 Overview
 Cells
 Segmentation and Reassembly
 Virtual Paths
 Physical Layers for ATM
 ATM in the LAN
NETE0510: Communication Media and Data
Communications
39
ATM in the LAN
 ATM can be used in LANs  a replacement for
Ethernet and 802.5
 Its popularity can be attributed to two main
factors:
ATM is a switched technology, whereas Ethernet and
802.5 were originally envisioned as shared-media
technologies
ATM was designed to operate on links with speeds of
155 Mbps and above, compared to the original 10 Mbps
of Ethernet and 4 or 16 Mbps of token rings
NETE0510: Communication Media and Data
Communications
40
ATM in the LAN (cont’d)
 Switched networks have a big performance advantage
over shared-media networks
 A single shared-media network has a fixed total bandwidth
that must be shared among all hosts, whereas each host gets
its own dedicated link to the switch in a switched network
 However, bridge that connects a number of sharedmedia network together is also a switch, giving dedicated
access
 Also when ATM were launched, high-speed Ethernet
became available. It speed began to approach that of
ATM
NETE0510: Communication Media and Data
Communications
41
ATM in the LAN (cont’d)
 One advantage of ATM over Ethernet that remains is the
lack of distance limitation for ATM links. Also high-speed
ATM links (e.g. 622 Mbps) became available.
 This made ATM fairly popular for the high-performance
“backbone” of larger LANs
 One common configuration is to connect hosts to
Ethernet switches, which in turn could be interconnected
by ATM switches
 High-performance servers might also be connected
directly to the ATM switch
NETE0510: Communication Media and Data
Communications
42
ATM in the LAN
NETE0510: Communication Media and Data
Communications
43
ATM in the LAN
 One significant problem with running ATM in a LAN is
that it doesn’t look like a “traditional” LAN
 Because most LANs are shared-media networks (i.e.
every node on the LAN is connected to the same link), it
is easy to implement broadcast and multicast
 Many protocols that people depend on in their LANs, e.g.
ARP, depend in turn on the ability of the LAN to support
broadcast and multicast
 However, ATM is not a shared-media network.
 How do we can broadcast to all nodes on an ATM LAN if
don’t know all their addresses and set up VCs to all of
them?
NETE0510: Communication Media and Data
Communications
44
ATM in the LAN (cont’d)
 Two possible solutions
 Redesign the protocols that make assumption about
LANs that are not in face true for ATM
 ATMARP does not depend on broadcast
 Make ATM behave more like a shared-media LAN 
supporting broadcast and multicast without losing the
performance advantages of a switched network
 LAN Emulation or LANE: aim to add functionality to ATM
LANs so that anything that runs over a shared-media LAN
can operate over an ATM LAN
NETE0510: Communication Media and Data
Communications
45
ATM in the LAN (cont’d)
 One aspect of LANE that can be confusing is the variety
of different addresses and identifiers that are used
 ATM devices must have ATM address used when
signaling to establish a VC
 Different from IEEE 802 MAC address used in Ethernet and
token rings
 LANE does not actually change functionality of ATM
switches, but add functionality to the network through the
additional of a number of servers.
 Devices connecting to the ATM network – hosts, bridges, routers
– are referred to as LAN emulation clients (LECs).
NETE0510: Communication Media and Data
Communications
46
ATM in the LAN (cont’d)
NETE0510: Communication Media and Data
Communications
47
LAN Emulation
 The servers that are required to build an emulated LAN
are:
 The LAN emulation configuration server (LECS)
 The LAN emulation server (LES)
 The broadcast and unknown server (BUS)
 These servers can be physically located in one or more
devices
 LECS and LES primarily perform configuration functions
 BUS has a central role in making data transfer in an ATM
network resemble that of a shared-media LAN
NETE0510: Communication Media and Data
Communications
48
LAN Emulation (cont’d)
LES
H1
BUS
H2
NETE0510: Communication Media and Data
Communications
49
Broadcasting in LAN Emulation
1. LECS enables a newly attached or rebooted LAN emulation client to
get some essential information
2. First, the client must find the LECS, which may use a well-known,
predefined VC that is always set up. Alternatively, the client must
have prior knowledge of the ATM address of the LECS so it can set
up a VC to it
3. Once connected, the client provides the LECS with its ATM address
4. LECS responds by telling the client
 what type of LAN is being emulated (Ethernet or token ring)
 What the maximum packet size is
 The ATM address of the LES. One LES might support many separate
emulated LANs
5. The client signals for a connection to the LES whose ATM address is
just learned.
6. Once connected to the LES, the client registers its MAC and ATM
addresses with the LES. The LES provides the client with the ATM
address of the BUS
NETE0510: Communication Media and Data
Communications
50
Broadcasting in LAN Emulation (cont’d)
7. The BUS maintains a single point-to-multipoint VC that connects it to
all registered clients. BUS and multipoint VC are crucial to LANE:

Enable the broadcast capability of traditional LAN to be emulated
in a VC environment
8. Once LEC has the ATM address of the BUS, it signals for a
connection to the BUS
9. The BUS adds the LEC to the point-to-multipoint VC


Now LEC is ready to participate in data transfer
BUS is the place to send any packet that needs to be
broadcast to all clients on the LAN, not efficient to deliver
unicast packets
NETE0510: Communication Media and Data
Communications
51
Unicasting in LANE
 Assume that a host has a packet to send to a particular MAC
address
 In traditional LAN, simply send packets over on the wire. It will be
picked up by intended recipient
 But in LANE, packet needs to be delivered over a VC. The sending
host also does not know the ATM address of the recipient, which is
required to set up a VC
 The host performs the following steps:
1. It sends the packet to the BUS, which it knows can deliver the
packet to the destination using its point-to-multipoint VC
2. It sends an “address resolution” request to the LES, of the
form “what ATM address correspond to this MAC address?”
NETE0510: Communication Media and Data
Communications
52
Unicasting in LANE (cont’d)



Since all clients should have registered their MAC
and ATM addresses with the LES, the LES should
be able to answer the query and provide an ATM
address to the client
The client now can signal for a VC to the recipient,
which may use to forward subsequent frames to the
destination
The reason for using BUS to send the first packet is
to minimize delay, since it may take some time to get
a response from the LES and establish a VC
NETE0510: Communication Media and Data
Communications
53
Questions?
Next Lecture
Wireless LANs and
Cellular Networks
NETE0510: Communication Media and Data
Communications
54