Transcript The Transport Layer

Chapter 6
The Transport Layer
2010
1
Services Provided to the Upper Layers
•connection-oriented and connectionless
•very similar to the network layer
•transport layer makes it possible for the transport service to
be more reliable than the underlying network service.
•Furthermore, the transport service primitives can be designed
to be independent of the network service primitives, which
may vary considerably from network to network.
•Dominant for network layer: IP thus connectionless
•for transport layer:
•TCP connection oriented
•UDP for connectionless: IP with a little extra
2010
2
Transport Service Primitives
2010
3
Berkeley Sockets
SOCKET: Create a new communication end point. The parameters specify the
addressing format, the type of service (e.g. reliable byte stream) and the protocol. It
returns an ordinary file descriptor for use in succeeding calls, (same as OPEN).
BIND: Binds a local address to a socket.
LISTEN: Allocates space to queue incoming calls for the case that several clients try
to connect at the same time, it does not block.
ACCEPT: Blocks a server until a connect request TPDU from a client arrives. A new
socket with the same properties as the original LISTEN one is then created by the
transport entity and a file descriptor for it is returned. The server can then fork off a
process or tread to handle the connection on the new socket and go back to waiting for
the next connection on the original socket by issuing a new ACCEPT primitive.
CONNECT: Issued by a client after it has created a socket, it blocks the client and
actively starts the connection process. When it completes (the appropriate TPDU is
received ), the client process is unblocked and the connection is established.
SEND: Sends some data over the connection.
RECEIVE: Receives some data from the connection
CLOSE: Symmetric connection release.
2010
4
Transport vs Data link Protocol
•Both have to deal with error control, sequencing, flow control and other
issues.
•In the subnet a packet might be stored for a number of seconds and then
delivered later. This can sometimes be disastrous, specially during
connection and disconnection, and requires the use of special protocols.
•The large and dynamically varying number of connections in the
transport layer, requires a different buffering and flow control approach
than in the data link layer. The idea of dedicating many buffers to each
connection in the transport layer is not attractive.
2010
5
Flow control
•In data link and transport layer a sliding window or other
scheme is needed on each connection to keep a fast
transmitter from overrunning a slow receiver.
•A router has relatively few lines whereas a host may have
many open connections.
•The optimum tradeoff between source and destination
buffering depends on the reliability of the connection and the
type of traffic.
•For low bandwidth, bursty traffic (like produced by an
interactive terminal) it is better to buffer at the sender. For
high bandwidth, smooth traffic (like a file transfer) it is better
to buffer at the receiver.
•Dynamically adjustment of buffer allocation is needed.
2010
6
Decouple buffering
•A general way to manage dynamic buffer allocation is to
decouple the buffering from the acknowledgments, in
contrast to the sliding window protocol of the data link layer.
•Initially the sender requests a certain amount of buffer space
on the receiver side, based on its perceived needs. The
receiver grants as many as it can afford.
•Every time the sender transmits data, it must decrement its
allocation, stopping all together if it reaches 0.
•The receiver separately piggybacks both acks and the
amount of available buffer space onto the reversed traffic.
2010
7
UDP
•UDP (User Datagram Protocol) is just basically IP with a short header
added. The port numbers indicate the sending and receiving transport
end points. When a UDP packet arrives, its payload is send to the
process attached to the destination port.
•The checksum is optional and stored as 0 if not computed, a calculated
0 checksum is stored as all 1s. UDP does not do flow control, error
control or retransmission upon receipt of a bad datagram. All of that is
up to the user process.
2010
8
Internet Transport Protocol: TCP
•TCP (Transmission Control Protocol) provides a reliable
byte stream over an unreliable internetwork.
•TCP accepts user data streams from local processes, breaks
them up into pieces not exceeding 64K (usually about 1460
bytes to fit in a single Ethernet frame) and sends each piece
as a separate IP datagram.
•The receiver side gives IP datagrams containing TCP data to
its TCP entity, which reconstructs the original byte streams.
•IP gives no guarantees that datagrams will be delivered
properly, so it is up to TCP to time out and retransmit. Also IP
datagrams might be delivered in the wrong order, it is up to
TCP to rearrange them in the proper sequence.
2010
9
The TCP Service Model
Port
21
23
25
69
79
80
110
119
Protocol
FTP
Telnet
SMTP
TFTP
Finger
HTTP
POP-3
NNTP
Use
File transfer
Remote login
E-mail
Trivial File Transfer Protocol
Lookup info about a user
World Wide Web
Remote e-mail access
USENET news
Port numbers below
1024 are called
well-known ports
and are reserved for
standard services.
The TCP service is obtained by having both the sender and receiver create end points,
called sockets. Each socket has a socket number (address) consisting of the IP address
and a 16 bit number local to that host, called a port.
Two or more connections may terminate at the same socket. Connections are
identified by the socket identifiers at both ends, that is (socket1, socket2).
All TCP connections are full-duplex (traffic can go in both directions) and point-topoint (each connection has exactly 2 end points). Multicasting or broadcasting are not
supported.
2010
10
Byte stream
A TCP connection is a byte stream, not a message stream. For example, if the
sending process does four 512 byte writes to a TCP stream, these data may be
delivered to the receiving process as four 512 byte chunks, two 1024 bytes chunks,
one 2048 byte chunk or some other way. This is analogous to UNIX files.
TCP may send user data immediately or buffer it, in order to send larger segments to
the network layer (larger IP datagrams.)
Applications might use the PUSH flag to indicate that TCP should send the data
immediately. The application can also send urgent data (e.g. when an interactive
user hits the CTRL-C key), TCP then puts control information in its header and the
receiving side interrupts (gives a signal in UNIX terms) to the application program
using the connection.
2010
11
TCP Segment Header
2010
12
The TCP Checksum
•It checksums the header, the data and a conceptual pseudo header.
•The pseudo header contains IP addresses and thus violates the protocol
hierarchy. It helps to detect wrongly delivered packets.
•The checksum is a simple one, just adding the 16-bits words in 1's
complement and then take the 1's complement of the sum. The
receiving side should find a 0.
2010
13
TCP Connection Establishment
A 3-way handshake. CR (connection request) and REJECT are
indicated by the SYN and FIN bits in a segment header.
The initial sequence numbers are chosen in a special way, ensuring that
no delayed or duplicated older request can fake a connection.
2010
14
TCP Connection Management Modeling
There are 11 states
used in the TCP
connection
management finite
state machine. Data
can be send in the
ESTABLISHED and
the CLOSE WAIT
states and received in
the ESTABLISHED
and FINWAIT1
states.
Releasing a TCP connection is symmetric. Either part can send a TCP segment with
the FIN bit set, meaning it has no more data to send. When the FIN is acked, that
direction is shut down, but data may continue to flow in the other direction.
It is not really fool proof, but it works.
2010
15
TCP Finite state Machine
•Each line is marked by an
event/action pair.
•The event can either be a userinitiated system call (CONNECT,
LISTEN, SEND or CLOSE), a
segment arrival (SYN, FIN, ACK
or RST), or a timeout.
•For the TIMED WAIT state the
event can only be a timeout of
twice the maximum packet length.
•The action is the sending of a
control segment (SYN, FIN or
RST) or nothing.
•The time-outs to guard for lost
packets ( e.g. in the SYN SENT
state) are not shown here.
2010
16
TCP Window management
Sender buffering:
Senders are not required to
transmit data as soon as they
come in from the application.
Usually Nagle's algorithm is
used. When data come into the
sender one byte at a time (e.g.
on a Telnet connection), just
the first byte is send and the
rest is buffered until the
outstanding byte is acked.
Sometimes it is better to
disable this, e.g. when mouse
movements are sent in XWindows applications.
2010
17
Silly window syndrome
•This occurs when the sender transmit data in large blocks, but an
interactive receiver application reads data 1 byte at a time.
•The receiver continuously gives 1 byte window updates and the sender
transmits 1 byte segments.
•Clark's solution is to let the receiver only send updates when it can
handle the maximum segment size it advertised when the connection
was established or when its buffer is half empty.
•Receivers are also not required to send acks and window updates
immediately.
•Many implementations delay them for 500 msec in the hope to
piggyback them.
2010
18
TCP Congestion Control
•It is assumed that time-outs are caused by network or receiver
congestion as lost packets are rare these days.
(Wireless networks are now an exception)
•Each sender maintains two windows:
•the window the receiver has granted (indicating the receiver
buffer capacity)
•the congestion window (indicating the network capacity).
•The number of bytes that may be sent is the minimum of the 2
windows.
2010
19
TCP slow start
The congestion window is
doubled on each packet
successfully sent (an ack
received before the
timeout). This exponential
increase (called the slow
start) continues until the
threshold (initially 32K)
is reached, after which the
increase is linearly.
When a timeout occurs, the threshold is set to half the current
congestion window and the slow start is repeated. If an ICMP source
quench packet comes in, it is treated the same way as a timeout.
2010
20
TCP Timer Management
For each segment the round trip time M is measured and the estimates of the mean and
mean deviation are updated as:
RTT = ß*RTT + (1-ß)*M
D = ß*D + (1-ß) * |RTT-M|
with ß a smoothing parameter, typically 7/8. The timeout is then set to: RTT + 4 D
2010
21
Improvements
•
•
•
•
•
Change from the original Go-Back-N to an more Selective Repeat
type of retransmit of lost segments:
• Buffering out-of-order segments in the receiver
• Sender detects missing segments by looking at duplicate ACK’s
(NACK’s can not be used) and retransmits before the timer
expires.
Improved congestion control:
• look also to duplicate ACK’s to lower the congestion window to
half its value
• start from that value with linear increase
versions to handle HTTP (www) trafic efficiently
versions for high speed links to increase the throughput
On a wireless network if a packet is lost, it is usually not due to
congestion, but due to transmission "noise". TCP should not slow
down, but retransmit as soon as possible.
2010
22
Wireless TCP
• Using an indirect TCP solution is a possibility. But the
acknowledgment, returned from the base station to the sender, does
not indicate that the mobile host has received the data.
• Another possibility is to add a snooping agent on the base station. It
watches the interchange between base station and mobile host and
performs retransmissions (and interception of duplicated
acknowledgments) on that part alone. However there is still the
possibility that the sender times out and starts it congestion control.
2010
23
Real-Time Transport Protocol
•RTP is intended for real-time
multimedia applications
•Its basic function is to
multiplex several real-time
data streams into a single
stream of UDP packets
•It can be regarded as a
transport sublayer
•Used for e.g. radio, telephony, music-on-demand, videoconferencing,
video-on-command, etc.
•It may contain for example a video stream and 2 audio stream for
stereo or sound tracks in 2 languages.
2010
24
RTP header
•Each payload may contain multiple samples, they may be coded any
way the user wants. The payload type field indicates which one is used.
•The timestamp gives the time of each sample relative to the first sample.
•Synchronization source identifier: uniquely identifies the synchronizing
source of a stream
•Contributing source identifier: uniquely identifies the contributing
streams (There are CC of them)
2010
25
some RTP audio/ video formats
2010
26
Real Time Control Protocol
•RTCP can be used to handle synchronization between the
sources.
•further it gives feedback on delay, jitter, bandwidth and
congestion.
•The encoding process can use it to increase the data rate and
give better quality if the network is functioning well,
•otherwise it can cutback in data rate and hence reduce the
quality.
•note that missing RTP packets are usually not requested but
are to some extend compensated for in the used encoding
method.
2010
27
Real-Time Streaming Protocol
RTSP
controls the
transmission
of a video
stream.
It does not
control the
stream itself.
2010
28