1. CSC 336 Data Communications and Networking Lecture 7b: Local Area Networking (Token Ring – 802.5) Dr. Cheer-Sun Yang Spring 2001

3. offices

5. Topologies Bus: A single communication line, typically a twisted pair, coaxial cable, or optical fiber, represents the primary medium. Ring: packets can only be passed from one node to it’s neighbor. Star: A hub or a computer is used to connect to all other computers. Tree: no loop exists (logical connection).

6. Token Passing Token Ring (802.5) : P. 183, Section 6.3 Token Bus (802.4) : P. 186, Section 6.4

8. Token Passing The difficulty with many networks is that no central control or authority makes decisions on who sends when. Token passing is designed to deal with this issue and hopefully the link utilization can be increased.

9. Token Passing In order to send, a station must obtain an admission pass, called a token. In a token ring, the token is passed from one station to another. When a station does not need it, it simply passes it on. Token ring network must pass the token orderly to it’s neighbor. Token bus network can pass a token to any other station directly.

10. Token Passing However, a token bus network cannot be added as simply as with the CSMA/CD bus. All stations must know who and where its neighbor is in a token bus.

11. 6.3 Token Ring: IEEE 802.5 Each repeater connects to two others via unidirectional transmission links Single closed path Data transferred bit by bit from one repeater to the next Repeater regenerates and retransmits each bit Repeater performs data insertion, data reception, data removal Repeater acts as attachment point Packet removed by transmitter after one trip round ring

12. Token Ring (802.5) MAC protocol –Small frame (token) circulates when idle –Station waits for token –Changes one bit in token to make it SOF for data frame –Append rest of data frame –Frame makes round trip and is absorbed by transmitting station –Station then inserts new token when transmission has finished and leading edge of returning frame arrives –Under light loads, some inefficiency –Under heavy loads, round robin

13. Dedicated Token Ring 1997 update to IEEE 802.5 Central hub Acts as switch Full duplex point to point link Concentrator acts as frame level repeater with immediate access possible No token passing if using a switch to connect all stations

14. 802.5 Physical Layer Data Rate416100 Medium UTP,STP,Fiber Signaling Differential Manchester Max Frame45501820018200 Access ControlTP or DTRTP or DTRDTR Note: 1Gbit in development

15. Ring Repeater States

16. Listen State Functions Scan passing bit stream for patterns –Address of attached station –Token permission to transmit Copy incoming bit and send to attached station, while forwarding each bit Modify bit as it passes –e.g. to indicate a packet has been copied (ACK)

17. Transmit State Functions Station has data Repeater has permission May receive incoming bits –If ring bit length shorter than packet Pass back to station for checking (ACK) –May be more than one packet on ring Buffer for retransmission later

18. Bypass State Signals propagate past repeater with no delay (other than propagation delay) Partial solution to reliability problem (the failure of one station can cause the network failure). Improved performance

19. Ring Media Twisted pair Baseband coaxial Fiber optic Not broadband coaxial –Would have to receive and transmit on multiple channels, asynchronously

20. Two observations 1.1. Ring contention is more orderly than 2. with an Ethernet. No wasted bandwidth.

21. Two observations 2. The failure of one station can cause network failure. More discussion will be provided in next slide.

22. Advantage of Token Ring The flexible control over access that it provides. The access is fair. It is easy to provide priority and guaranteed bandwidth services.

23. Disadvantage of Token Ring Token maintenance requires extra work. Loss of token prevents further utilization of the ring. Duplication token can disrupt the operation. A monitor station is required. It becomes a crucial point for a single point failure.

24. Potential Ring Problems Break in any link disables network Repeater failure disables network Installation of new repeater to attach new station requires identification of two topologically adjacent repeaters Method of removing circulating packets required –With backup in case of errors Mostly solved with star-ring architecture (the wire center approach).

25. Network Failure Problem The failure of one station can cause network failure: This problem can be solved by using a wire center (Fig. 6.11). Instead of connecting neighboring stations directly, they all communicate through a wire center. The wire center contains a bypass relay. If a station fails, the bypass relay will allow a frame to bypass the station. This architecture is called a Star Ring Architecture.

27. Star Ring Architecture Feed all inter-repeater links to single site –Concentrator –Provides central access to signal on every link –Easier to find faults –Can launch message into ring and see how far it gets –Faulty segment can be disconnected and repaired later –New repeater can be added easily –Bypass relay can be moved to concentrator –Can lead to long cable runs Can connect multiple rings using bridges

28. Timing Jitter Clocking included with signal –e.g. differential Manchester encoding –Clock recovered by repeaters To know when to sample signal and recover bits Use clocking for retransmission –Clock recovery deviates from midbit transmission randomly Noise Imperfections in circuitry Retransmission without distortion but with timing error Cumulative effect is that bit length varies Limits number of repeaters on ring

29. Solving Timing Jitter Limitations Repeater uses phase locked loop –Minimize deviation from one bit to the next Use buffer at one or more repeaters –Hold a certain number of bits –Expand and contract to keep bit length of ring constant Significant increase in maximum ring size

30. A Closer Look on Token Passing 20 stations, each separated by 10 meters, for a total ring length of 200 meters Transmission at 4Mbps, or one bit every 0.25  sec Propagation speed of 2*10 8 m/s How long does it take to travel around the ring once? - 1  sec How many bits can a station send in 1  sec?

31. A Closer Look on Token Passing A: 4 bits. To get more bits on the ring, each station delays one bit-time allowing it to examine each bit before deciding whether to copy it or repeat it. (one bit-time = 0.25  sec) How many bits can a station send then?

33. Token and Frame Formats Start Delimiter (SD), End Delimiter (ED): 1 octet Access Control (AC) : 1 octet, 3 priority bits, 1 token bit, 1 monitor bit, 3 reserved bits. Token bit determines the frame type, i.e. token frame or data frame. Priority bit can be used to set the token’s priority. Monitor bit and reserved bits are used for ring maintenance.

34. Token and Frame Formats Frame Control (FC): used to distinguish control frame from data frame. Frame Status(FS): 1 octet (acxxacxx) a: address recognized bit, c: frame copied bit, x: undefined bit. –a = 0, c = 0: dest not present or not power up –a = 1, c = 0: dest present but frame is not accepted –a = 1, c = 1: dest present and frame copied.

35. Token Ring Operation

36. Priority Scheme 1.A station having a higher priority frame to transmit than the current frame can reserve the next token for its priority level as the frame passes by. 2.When the next token is issued at a station A, it will be at the reserved priority level. The station reserving the token can use this token to transmit data frame. 3.The station A is responsible to down-grade the priority of the token later.

37. Priority Scheme A sends a frame to B at priority 0. When the frame passes by D, D makes a reservation at priority 3. When the token is sent back to A, A changes the priority to 3 and issues a new token. D can use this token to send a frame to any station. After the data is seized by the destination and the token is passed back to A, A is responsible for changing the priority back to 0. (Why A?)

38. Priority Scheme In this case, station A is called a stacking station. When it generates a new token with a higher priority, it also keeps the old priority in a stack locally. Thus, a stacking station is the only station in which the old priority is kept.

39. Reserving and Claiming Tokens C B D A token

40. Reserving and Claiming Tokens C B D A Station A requests the token and sends its data to D

41. Reserving and Claiming Tokens C B D A Station C can reserve the next open token By entering its priority code in the AC field.

42. Reserving and Claiming Tokens C B D A Station D copies the frame and sends the data back to the ring.

43. Reserving and Claiming Tokens C B D A Station A receives the frame and releases the token

44. Reserving and Claiming Tokens C B D A Station C can send its data now.

46. Priority Scheme

47. Time Limits Token holding time: the time duration a station is allowed to hold the token Token rotation time: the total time a token is allowed to rotate around the ring. TRT >= N * THT

48. Ring Maintenance Things can go wrong. For example: 1.A station sends a short frame over a long ring and subsequently crashes. It is not able to drain the token. This frame is called an orphan frame. 2.A station receives a frame or token crashes before it can send it. Now there is no token circulating. 3.Line noise damages a frame.

49. Ring Maintenance Some problems can be handled by giving one of the stations a few different responsibilities and designating it a monitor station. 1.When a monitor station receives a frame, it sets the monitor bit to 1. If the frame is received the second time and the monitor bit is still set to 1, the monitor station deletes the frame.

50. Ring Maintenance 2. The monitor station also detect a lost token using a built-in timer which is determined based on the length of the ring, number of stations, and maximum frame size. Whenever the monitor sends a frame or token, it starts the timer. If the monitor does not receive another frame or token before the timer expires, it assumes that the token is lost. It then creates another one.

51. Ring Maintenance Some problems cannot be solved even with a monitor station. For example, what if the malfunction station is the monitor station? What if a break in the ring causes a lack of tokens? Sending new ones does nothing to correct the problem. These problems are handled using control frames.

52. Ring Maintenance Some example control frames: Claim token frame – for submitting bids to elect a monitor station. Active monitor present (AMP) frame – to notify others that a monitor station has been produced. Standby monitor present (SMP) – frame. Beacon frame – to inform stations that a problem has occurred and the token-passing protocol has stopped.

54. Ring Efficiency T 1 = time to send a frame T 2 = time to send a token

55. Other Ring Networks: FDDI 100Mbps LAN and MAN applications Token Ring

57. Encoding Schemes 4B/5B-NRZI MLT-3

59. FDDI MAC Frame Format

61. FDDI MAC Protocol As for 802.5 except: Station seizes token by aborting token transmission Once token captured, one or more data frames transmitted New token released as soon as transmission finished (early token release in 802.5)

62. FDDI Operation

63. FDDI Physical Layer MediumOptical FiberTwisted Pair Data rate100100 Signaling4B/5B/NRZIMLT-3 Max repeaters100100 Between repeaters2km100m

64. LAN Generations First –CSMA/CD and token ring –Terminal to host and client server –Moderate data rates Second –FDDI, FDDI II –Backbone –High performance workstations Third –ATM –Aggregate throughput and real time support for multimedia applications

65. Third Generation LANs Support for multiple guaranteed classes of service –Live video may need 2Mbps –File transfer can use background class Scalable throughput –Both aggregate and per host Facilitate LAN/WAN internetworking

66. ATM LANs Asynchronous Transfer Mode Virtual paths and virtual channels Preconfigured or switched Gateway to ATM WAN Backbone ATM switch –Single ATM switch or local network of ATM switches Workgroup ATM –End systems connected directly to ATM switch Mixed system

67. Example ATM LAN

68. ATM LAN HUB

69. Compatibility Interaction between end system on ATM and end system on legacy LAN Interaction between stations on legacy LANs of same type Interaction between stations on legacy LANs of different types

70. Fiber Channel - Background I/O channel –Direct point to point or multipoint comms link –Hardware based –High Speed –Very short distance –User data moved from source buffer to destiation buffer Network connection –Interconnected access points –Software based protocol –Flow control, error detection &recovery –End systems connections

71. Fiber Channel Best of both technologies Channel oriented –Data type qualifiers for routing frame payload –Link level constructs associated with I/O ops –Protocol interface specifications to support existing I/O architectures e.g. SCSI Network oriented –Full multiplexing between multiple destinations –Peer to peer connectivity –Internetworking to other connection technologies

72. Fiber Channel Elements End systems - Nodes Switched elements - the network or fabric Communication across point to point links

73. Fiber Channel Network

74. Fiber Channel Protocol Architecture (1) FC-0 Physical Media –Optical fiber for long distance –coaxial cable for high speed short distance –STP for lower speed short distance FC-1 Transmission Protocol –8B/10B signal encoding FC-2 Framing Protocol –Topologies –Framing formats –Flow and error control –Sequences and exchanges (logical grouping of frames)

75. FC-3 Common Services –Including multicasting FC-4 Mapping –Mapping of channel and network services onto fiber channel e.g. IEEE 802, ATM, IP, SCSI Fiber Channel Protocol Architecture (2)

76. Wireless LANs IEEE 802.11 Basic service set (cell) –Set of stations using same MAC protocol –Competing to access shared medium –May be isolated –May connect to backbone via access point (bridge) Extended service set –Two or more BSS connected by distributed system –Appears as single logic LAN to LLC level

77. Types of station No transition –Stationary or moves within direct communication range of single BSS BSS transition –Moves between BSS within single ESS ESS transition –From a BSS in one ESS to a BSS in another ESS –Disruption of service likely

78. Wireless LAN - Physical Infrared –1Mbps and 2Mbps –Wavelength 850-950nm Direct sequence spread spectrum –2.4GHz ISM band –Up to 7 channels –Each 1Mbps or 2Mbps Frequency hopping spread spectrum –2.4GHz ISM band –1Mbps or 2Mbps Others under development

79. Media Access Control Distributed wireless foundation MAC (DWFMAC) Distributed coordination function (DCF) –CSMA –No collision detection Point coordination function (PCF) –Polling of central master

80. 802.11 MAC Timing

81. Reading Chapter 6: 6.1-6.5