Gigabit Token Ring - Case Study Example

Gigabit Token Ring of Learning Introduction Token ring is a LAN technology that makes use of tokens in data transmission over a ring topology. It was developed originally under IBM standards after which the IEEE was involved in setting the second standard designated as 802.5. There existed some difference under the two standards in that the number of devices in type 1 shielded cable was reduced to 250 from a maximum of 260 (Miller & Cummins, 2000). To date, four token rings standards have been launched in the industry with varying success. These are; 4mbps, 16mbps, 100mbps and the gigabit token ring. It is possible to make use either of the first two token rings in the same equipment. The drawback to this functionality is that the introduction of a new device to such network brings problems to the entire ring (Carlo, 1998). Background The token ring was first initiated by IBM in the early 80s in their research facility located in Zurich. In 1985, IBM launched its fist token-ring product as a form of adapter that could be used in their original personal computers. In the following year, in collaboration with Texas Instruments, they jointly developed a chipset to help other companies to develop devices that are compatible with the token ring. The year 1989 saw IBM introduce an improved version of their earlier token ring model which had a speed of 4 Mbit/sec. The improved version had a speed of 16 Mbit/sec and in line with this development; the IEEE 802.5 standard was extended to cover it (Muller, 2003). In the year 1994, the leading suppliers of token ring created the Alliance for Strategic Token-ring Advancement and Leadership (ASTRAL) whose main mission was increase the speed of Token-Ring technology to counter the ever rising popularity of Ethernet technology. The members of this group included: ACE/North Hills, 3Com, Bytex, Bay Networks, Cabletron, Chipcom , Centillion, IBM, Hewlett-Packard, , Intel, Olicom ,Madge, Racore, Proteon, Texas Instruments, SMC, UB Networks., Xircom, and XPoint (Muller, 2003). In 1997, a revised version of 802.5 standards was developed which marked the introduction of Dedicated Token-Ring. It was a form of full duplex token ring that bypassed the usual protocol used by token ring in data transfer. A single station thus had the capability of sending and receiving data streams concurrently. This had the effect of doubling the transfer rate of any token ring such that a 4 Mbit/sec dedicated Token-ring station acquired an overall transfer rate of 8 Mbit/sec. Commonly used Token ring cabling They are mainly Type 1, type 6 and type 3. The first wiring done on token ring used the Type 1 (STP) shielded twisted-pair cable. Type 6 token ring cable is lighter and flexible than type 1 hence capable of taking twist and turns during network construction. Type 3 uses CAT5, CAT3 or CAT5Ee cabling with RJ-45 connectors (Held, 2000). How Token ring works Components and processes The Ring The topology used is the star-wired ring which is made up of the transmission medium and ring stations. Active Monitor It is found on each ring and its main task is to monitor the ring among other functions. Standby Monitor It checks for failures that may occur on the active monitor. Claim Token Process This process helps in the determination of the station which takes the role of the active monitor. Ring Purge Process This process facilitates the release of a new token as well as cleaning the ring. Beacon Process This process is initiated if the claim token process fails to take place. Neighbor Notification Process This process ensures that the upstream neighbor identity is available to the station before it (Miller, 2009). Error Detection and Reporting Soft errors are temporary in nature since they are rectified by error recovery procedures. Hard errors are usually permanent I nature affecting the cabling or equipments in the network. Phases for a station insertion into a token ring For a station to insert into a ring, it has to pass through the following phases (Habraken, 2004): Lobe Test: Phase 0 Monitor Check: Phase 1 Duplicate Address Check: Phase 2 Neighbor Notification: Phase 3 Request Initialization: Phase 4 Priority Operation: Phase 5 Token-ring formats Frame format A frame is the basic transmission unit in the operation of a token ring. This frame format can be used to transmit both MAC and LLC frames. In this frame there is no obligation for including the information or routing information field. The following fields make up the frame (Carlo, 1998); Starting delimiter – 1 byte Access control – 1 byte Frame control – 1 byte Dest. MAC Address – (6 byte) Source MAC Address – (6 byte) Routing information – (0-30 bytes) Information – (0-n bytes) Frame check sequence – (4 bytes) Ending delimiter - (1 byte) Frame status – (1 byte) Token format The right to transmit a frame from one station to the other is done in the form of token. Those stations that are capable of transmitting the message transform the token to the equivalent frame (Miller & Cummins, 2000). When the frame returns to its original station after a full cycle it is removed from the ring circulation and transmit a new token to undergo the same process as the previous one. The token formats carry the following fields; Starting Delimiter (1byte) Access Control (1byte) Ending delimiter (1byte) Abort Sequence Format In this format, a station transmitting the frame may abort the process at any time by initiating an abort sequence (Miller & Cummins, 2000). The preceding stations upon receiving it thus recognize it as invalid frame. It contains the following fields; Starting delimiter (1byte) Ending delimiter (1byte) Fields definitions Starting delimiter The abort sequence, token and frame format always starts with this field (Habraken, 2004). Its length is usually 1 byte and uses the symbols J and K to form a unique sequence that consists of code violations as shown in table 1 below; Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 J K 0 J K 0 0 0 table 1 J = Non Data “J” Symbol K = Non Data “K” Symbol 0 = Data “0” Symbol Access control This field is only found in tokens and frames and is usually 1 byte in length (Held, 2000). The parameters that are present are as shown in table 2 below; Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 P P P T M R R R Table 2 PPP stands for priority bits. These bits indicate which station should use the token in terms of determining the level of access priority. T stands for the token bit. It is usually set as b ‘1’ in frames and b‘0 in tokens.’ M stands for the monitor bit. Hinders a given frame from endlessly circulating the ring and also deny access a frame with a priority which is more than b’000’. This is done by setting the bit in question to b’1’ such that when it arrives at the active monitor it’s removed and a new token is released. RRR represent the reservation bits. Determines the priority level assigned to repeat token and frames in regard to stations priority access. b’000’ is the lowest priority level while the highest is b’111’. Frame control This field is usually 1 byte in length and helps in indicating the frame type (Held, 2000). Its component is as shown in table 3 below; Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 F F Z Z Z Z Z Z Table 3 FF stands for Frame Type Bits and indicates the frame type. For MAC frame it is b’00’ and for LLC Frame it is b’01’. ZZZZZZ represents the control bits. It is Present in MAC Frame to distinguish whether a frame should be regarded S “express” or”normal”. If the bits indicated are zeros it is a normal buffered frame and if its non-zero then it is an express buffered frame. Destination address It is mainly has a length of 6 bytes. It helps in the identification of those stations that should copy the frame (Habraken, 2004). The fields are as shown in table 4; Bit0 Bit1 Bit2 – Bit 47 I/G U/L Table 4 I/G - individual/group. The bit helps in the identification of an address as either group address (b’1’) or individual address (b’0’). U/L - Universal/Local. Helps in the identification if an address deserves to be locally administered (b’1’) or universally administered (b’0’). Source address This field is made up of six bytes that help in the identification of the initial station that generates the frame. It contains the following fields as shown in table 5; Bit0 Bit1 Bit2 – Bit 47 RII U/L Table 5 RII stands for routing information indicator. This field will always carry an individual address which helps in the determining if the field with Routing Information can be found in the frame. If b’1’, it should be included in the frame and if b’0’ it should be excluded. U/L - Universal/Local. This bit possesses similar functions as those in Destination Address Field. It helps in indicating whether the address can be classified as locally administered (b’1’) or universally administered b’0’. Routing Information This field is applied during routing of frames in a network with multiple rings. This field is however omitted if a frame is not bound to leave the source ring (Miller & Cummins, 2000). The bit on routing information indicator signifies its presence if the source address field is b’1’. The RI has a permitted length of up to 30 bytes but only up to 18 is implementable. The designator fields present in routing information is as follows in table 6; Routing Control (2-bytes) Route Designator (2-bytes) Route Designator (2-bytes) ... Route Designator (2-bytes) Table 6 Routing control This field possesses the bit definitions as shown in table 7 below; Byte0 Byte1 Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 B B B L L L L L Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 D F F F r r r r Table 7 BBB stands for Broadcast Indicators. These bits provide an indication on how a frame will be broadcasted in case f multiple ring segments (Miller & Cummins, 2000). LLLLL stands for Length Bits. The RI field length represented in bytes is indicated using the 5 bits. D represents Direction Bits which ensures that the routing descriptor remains in the right order. FFF stands for Largest Frame Bits. The largest information under transmission between any two stations is specified by this field. rrrr stands for Reserved Bits. They are usually transmitted in zeros. Route Designator This field is composed of a ring number made of 12-bit and a bridge number made of 4-bits as shown in table 8; Byte0 Byte1 Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 RN (12bits) BN (4bits) RN stands for Ring Number BN stands for Bridge Number Information Field The data under transfer is located in this field. Information Field (0 to n bytes) Frame Check Sequence This field is composed of a 4-byte cyclical redundancy check (CRC) value tasked with error checking (Miller, 2009). Byte0 Byte1 Byte2 Byte3 FCS FCS = Frame Check Sequence Ending Delimiter It is usually one byte in length and is considered valid by the receiving station if the first six symbols as shown in table 9 are correctly received. Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 J K 1 J K 1 I E Table 9 J stands for Non Data "J" Symbol K stands for Non Data "K" Symbol 1 stands for Data "1" Symbol I stands for Intermediate Frame Bit which is transmitted as b1. E stands for Error Detected Bit. If it originates from abort sequence, frame or token, it is usually transmitted as b0. Frame Status This format in this field is as shown in table 10 below; Bit0 Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 A C r R A C r r Table 10 A & C stands for Address Recognized (A) and Frame Copied (C) Bits r stands for Reserved Bits. The development of gigabit project In the year 1997, some of the members of the High Speed Token-Ring Alliance (HSTRA) came up with the idea of improving the IEEE 802.5 standard to develop dedicated high-speed Token Ring that would manage a data transfer rate of above 100 Mbit/sec to a minimum of 1 gigabit. These members included Bay Networks 3Com, Madge Networks, IBM, Xylan, Olicom, and the UNH Interoperability Lab (Muller, 2003). The above efforts saw the emergence of draft encompassing the 802.5t standard which defined the proposed 100Mbit/sec Token Ring which was restricted to dedicated operations. IBM and Olicom were the first companies to introduce the 100 Mbi/sec Token-Rings products (Muller, 2003). A working group using the draft standard 802.5v in the early 2000 was later tasked in developing a Gigabit Token ring. Gigabit Token Ring features Large Frames Multiple Access Priorities No head-of-line blocking Source Routing Load sharing Resilience on redundant links Properties and features of Gigabit token ring Followed the IEEE 802.5v specification Made use of logical physical ring topology Speeds were assigned as 1000Mbps or more (Miller, 2009) Ensured the data transfer is collision-free Utilizes an MSAU (multi-station access unit) rather than a switch or hub Possesses a high overhead Merits of a gigabit token ring Provide a speed equal or greater than 1 gigabit. Enhanced the IEEE 802.5t protocol High support for the emerging application that is bandwidth-intensive Conclusion The token ring was mainly a success in the 80s and 90s. In this period, IBM accorded its users maximum support in the 4 Mbit/sec and 16 Mbit/sec versions. The development of token rings with a capacity of 100 Mbit/sec was a major breakthrough for the ASTRAL alliance. However, after this move HSTR was on the decline due to competition from high speed Ethernet which was much cheaper to install for the end user (Muller, 2003). Due to low penetration of this technology in the market, most of the members in the ASTRAL were unable to push the gigabit token ring project to its full potential. Only three members were left in the project of developing Gigabit token ring, theses are Olicom, IBM and Madge (Muller, 2003). The reduction in customer base due to heavy competition from fast Ethernet ultimately made the investment on gigabit token ring unworthy while for the three major players in ring technology. Despite the gigabit token ring being standardized in the year 2001, it never achieved full implementation. References Carlo, J. T. (1998). Understanding token ring protocols and standards. Boston: Artech House. Habraken, J. W. (2004). Absolute beginners guide to networking. Indianapolis, Ind: Que. Held, G. (2000). Network design: Principles and applications. Boca Raton: Auerbach. Miller, P. (2009). TCP/IP: The ultimate protocol guide. Boca Raton, Fla: BrownWalker Press. Miller, P., & Cummins, M. (2000). LAN technologies explained. Boston: Digital Pres Muller, N. J. (2003). LANs to WANs: The complete management guide. Boston: Artech House. Read More