1. Packet Switching Networks & Frame Relay
Chapter 4
Packet Switching Networks
&
Frame Relay

2. Introduction Packet-Switching Frame Relay Networks Switching Technique
Routing
X.25
Frame Relay Networks
Architecture
User Data Transfer
Call Control
Chapter 4 Frame Relay

3. Introduction - Taxonomy
Communication Networks
Circuit -Switched
Packet -Switched
FDM
TDM
Datagram
Virtual Circuit
The Internet
(TCP/IP)
Frame Relay
ATM
Chapter 4 Frame Relay

4. Circuit-Switching
Historically – long-haul telecom networks designed for voice and/or constant bit rate applications
Network resources dedicated to one “call” after circuit setup
Shortcomings when used for data:
Inefficient (high idle time) for “bursty” sources
Constant data rate not appropriate for varied endpoint capabilities
Chapter 4 Frame Relay

5. Packet-Switching
Historically – network technology designed for general data communications
Basic technology is the same as in the 1970s
One of the few effective technologies for long distance data communications in use today
Frame relay and ATM are variants of packet-switching (using virtual circuits)
Advantages:
flexible, resource sharing, robust, responsive
Disadvantages:
Time delays in distributed network, overhead penalties
Need for routing and congestion control
Chapter 4 Frame Relay

6. Packet-Switching Data transmitted in short blocks, or packets
Packet length typically < 1000 octets
Each packet contains user data plus control info (routing)
Store and forward
Chapter 4 Frame Relay

7. Use of Packets
Chapter 4 Frame Relay

8. A Simple Switching Network
Chapter 4 Frame Relay

9. Advantages over Circuit-Switching
Greater line efficiency (many packets can go over shared link)
Data rate conversions
Non-blocking (e.g. no “busy signals”) under heavy traffic (but increased delays)
Each packet can be handled based on a priority scheme
Chapter 4 Frame Relay

10. Disadvantages relative to Circuit-Switching
Packets incur delay with every node they pass through
Q * (dprop + dtrans + dqueue + dproc)
Jitter: variation in end-to-end packet delay
Data overhead in every packet for routing information, etc
More processing overhead for every packet at every node traversed… circuit switching has little/no processing at each node
Chapter 4 Frame Relay

11. Switching Technique
Large are messages broken up into smaller “chunks,” generically called packets
Store and forward packet handling in core
Two approaches to switching data:
Datagram
Each packet sent independently of the others
No call setup
More reliable (can route around failed nodes or congestion)
Virtual circuit
Fixed route established before any packets sent
No need for routing decision for each packet at each node
Chapter 4 Frame Relay

12. Packet Switching: Datagram Approach
Advantages:
No call setup
Flexible routes
Reliability
Chapter 4 Frame Relay

13. Packet Switching: Virtual-Circuit Approach
Advantages:
Network services
sequencing
error control
Performance
Chapter 4 Frame Relay

14. Routing
Key function of any packet-switched network: forwarding packets to a destination
Adaptive routing, routes are adjusted based on:
Node/trunk failure
Congestion
Nodes (routers/switches) must exchange information about the state of the network
Chapter 4 Frame Relay

15. The Use of Virtual Circuits
Virtual end-to-end
circuits
Chapter 4 Frame Relay

16. X.25 First commercial packet switched network interface standard
Motivates discussion of frame relay and ATM design
X.25 defines 3 levels of functionality
L1 - Physical level (X.21, EIA-232, etc.): physical connection of a station to the link
L2 - Link/frame level (LAPB, a subset of HDLC): logical, reliable transfer of data over the physical link
L3 - Packet level: network layer, provides virtual circuit service to support logical connections between two subscriber stations (multiplexing)
Chapter 4 Frame Relay

17. User Data and X.25 Protocol Control Information
Virtual circuit id#
Sequence #s
3 bytes
 128 bytes
Flags, address, control, FCS
Link layer framing
Reliable physical transfer
Chapter 4 Frame Relay

18. X.25 Features Call control packets Multiplexing of VCs at layer 3
set up and tear down virtual circuits
use same channel and VC as data packets
Multiplexing of VCs at layer 3
Layers 3 (packet) and 2 (frame) both include extensive flow control and error control mechanisms
Processing
Overhead
(tproc)
at each node!
RESULT:
64kbps
Max. data rate
Chapter 4 Frame Relay

19. Frame Relay Networks
Most widely deployed WAN link-layer protocol in use today
Designed to eliminate much of the processing overhead in X.25
Designed to support “bandwidth on demand” for modern, bursty applications
Throughput is an order of magnitude higher than X.25
ITU-T Recommendation I.233 indicates effective rates of frame relay of up to 2 Mbps, but current practice is much higher (up to T-3 equivalent, or Mbps)
Chapter 4 Frame Relay

20. Frame Relay Networks Important Improvement over X.25:
Call control signaling is on a separate logical connection from user data
Multiplexing/switching of logical connections is at layer 2 (not layer 3)
No hop-by-hop flow control and error control; responsibility of higher layers
Frames sizes can vary (up to 9000 bytes), supporting all current LAN frame sizes
Direct support for TCP/IP packets, since no network layer redundancy
Chapter 4 Frame Relay

21. Comparison of X.25 and Frame Relay Protocol Stacks
Chapter 4 Frame Relay

22. Virtual Circuits and Frame Relay Virtual Connections
(a) X.25 Packet Switching
X.25 Packet-Switching
network
(b) Frame Relay
Frame Relay network
Chapter 4 Frame Relay

23. Frame Relay Architecture
X.25 has 3 layers: physical, link, network
Frame Relay has 2 layers: physical and data link (or LAPF)
LAPF core: minimal data link control
Preservation of order for frames
Small probability of frame loss
Chapter 4 Frame Relay

24. LAPF Core Frame delimiting, alignment and transparency
Frame multiplexing/demultiplexing
Inspection of frame for length constraints
Detection of transmission errors
Congestion control
Chapter 4 Frame Relay

25. LAPF-core Formats 10-bit address 23-bit address 16-bit address
Chapter 4 Frame Relay

26. User Data Transfer Frame
No connection control fields, which are normally used for:
Identifying frame type (data or control)
Sequence numbers, used for error/flow control
Implication:
Connection setup/teardown carried on separate channel
No flow and error control, must be handled by higher layer in protocol stack
Chapter 4 Frame Relay

27. Frame Relay Call Control
Details of call control depend on the context of its use
Assumes FR over ISDN
Generally simpler for point-to-point use
Data transfer involves:
Establish logical connection and assign a unique DLCI
Exchange data frames
Release logical connection
Chapter 4 Frame Relay

28. Frame Relay Call Control
4 message types needed
SETUP…request link establishment
CONNECT…reply to SETUP with connection accepted
RELEASE…request to clear (tear down) a connection
RELEASE COMPLETE… reply to SETUP with connection denied, or response to RELEASE
Chapter 4 Frame Relay