WO1998030060A1

WO1998030060A1 - Method using a standardized protocol for transmission of frames of data across a high speed bus

Info

Publication number: WO1998030060A1
Application number: PCT/US1997/024206
Authority: WO
Inventors: Nicholas Ilyadis; Richard W. Graham; Barry A. Spinney
Original assignee: Cabletron Systems, Inc.
Priority date: 1996-12-30
Filing date: 1997-12-30
Publication date: 1998-07-09
Also published as: AU5811798A

Abstract

Frames of data of arbitrary length and protocol are received by a switch from a transmitter and encapsulated with a standardized header and trailer to produce a number of small bursts which are transmitted on to a high speed bus. The standardized header provides, among other things, a way in which reassembly of the bursts at a receiver occurs. The trailer typically contains a cyclic redundancy check to ensure data integrity. The process is protocol-independent, i.e., the transmitter and receiver of the frames may be of similar or dissimilar protocols, such as Ethernet, Fiber Distributed Data Interface (FDDI), and Asynchronous Transfer Mode (ATM).

Description

METHOD USING A STANDARDIZED PROTOCOL FOR TRANSMISSION OF FRAMES OF UATA ACROSS A riiflH SPEED bϋS

Background of the Invention

The present invention relates to the field of computer networks, and more particularly, to a method using a standardized protocol for transmission of frames of data as bursts across a high speed bus.

As is well known in the field of computer networks, such as local area networks (LANs), frames of data typically flow from a source system through one or more bridges or switches, and out to a destination system. As is also known, frames of data flow in computer networks according to protocols defined by certain well established standards. For example, in Ethernet networks, frames of data adhering to an Ethernet protocol are reduced to bursts of data containing from 64 to 1518 bytes of data. Thus, frames of data in an Ethernet network will flow from a source system, through a switch, for example, and on to a destination station, as bursts containing from 64 to 1518 bytes of data.

In Fiber Distributed Data Interface (FDDI) networks, frames of data adhering to an FDDI protocol are reduced to bursts of data containing from 17 to 4,000 bytes of data. Thus, frames of data in an FDDI network will flow from a source system, through a switch, for example, and on to a destination system, as bursts containing from 17 to 4,000 bytes of data.

In Asynchronous Transfer Mode (ATM) networks, frames of data adhering to an ATM protocol are reduced to bursts of data containing fixed length cells of 56 bytes of data. Thus, frames of data in an ATM network will flow from a source system, through a switch, for example, and on to a destination system, as bursts containing fixed length cells of 56 bytes.

As is also known in the art, computer networks often include two or more switches. The 98/30060

switches are often connected to each by a high speed bus. For example, in an Ethernet network, a source system may send frames of data as a burst of 64 to 1518 bytes of data to a port on a first switch. The burst will then σavel out a second port of the first switch and on to a high speed bus, finally arriving at a first port of a second switch. The burst may then flow out a second port of the second switch and arrive at a destination station.

Most network switches and high speed buses work well when handling a single protocol. For example, switching in an Ethemet-to-Ethemet network, bursts travel from sources systems, through a switch, out on a bus, through another switch, and finally out to destination systems, all according to the Ethernet protocol. In similar fashion, bursts in an FDDI-to-FDDI network travel according to the FDDI protocol, while bursts in an ATM-to-ATM network travel according to the ATM protocol.

As is now well known in the an, networks are becoming more complex and involving the mixing of Ethernet, FDDI, and ATM protocols in a single network. What is needed is a standardized protocol in the switches that encapsulate a payload of data with a standard header which allows applications to distinguish their traffic on the high speed bus.

Summaiv of the Invention

In accordance with one embodiment of the present invention, a method of transmitting a plurality of bursts of data including the steps of providing a source node, the source node generating a first frame of data to be transmitted to a bus, providing a network switch, the switch including a plurality of ports, a first port connected to the source node, a second port connected 8/30060

to the bus, a data moving engine (DME) connected to the plurality of ports, the DME conditioning the first frame comprising the steps of receiving the first frame, appending a primary header to the first frame to create a primary frame, determining whether the length of the of the first frame is greater than 64 bytes, appending a secondary header and a trailer to the primary frame in response to the step of determining to create a secondary frame, transmitting the secondary frame on to the bus, and repeating the steps appending and transmitting until all bytes in the first frame are transmitted on to the bus.

Brief Description of the Drawings The novel features believed characteristic of the invention are set forth in the appended claims. The invention itself, however, as well as other features and advantages there of, will be best understood by reference to the detailed description of specific embodiments which follows, when read in conjunction with the accompanied drawings, wherein:

FIG. 1 is a block diagram illustrating an exemplary local area network (LAN); FIG.2 is a block diagram illustrating an exemplary switch;

FIG. 3A-3C are block diagrams illustrating the data structures according to the principles of the invention;

FIG.4 is a block diagram of an exemplary header according to the principles of the invention; FIG. 5 is a block diagram an exemplary trailer according to the principles of the invention; and /30060

FIG. 6 is a flowchart detailing the process performed by the Data Moving Engine (DME) in accordance with the invention.

Description of the Preferred Embodiment.. _ _

Referring to FIG. 1 , an exemplary local area netwo± (LAN) 10 is shown to include four nodes labeled as 12, 14, 16, and 18, respectively. The exemplary LAN 10 is also shown as including two switches labeled as 20 and 22, respectively. Switch 20 includes three ports labeled as 24, 26, and 28, respectively, while switch 22 is shown having three ports labeled as 30, 32, and 34, respectively. Switch 20 is shown to be connected to a high speed bus 36 via port 28, while switch 22 is shown connected to the high speed bus 36 via port 34.

An exemplary transmission of data from node 12 (source) to node 18 (destination) proceeds in the following manner. A frame of data leaves node 12 and travels into switch 20 via the port 24. While in the switch 20, the frame is analyzed and checked for a source and destination address. If the source address and the destination address are valid, the data leaves the switch 20 and is placed on the high speed bus 36 via port 28. The data on the high speed bus 36 arrives at switch 22 via port 34, and flows on to node 18 via port 32.

It will be appreciated that the exemplary LAN 10 may contain more than the number of nodes and switches illustrated. In addition, in the exemplary transmission explained above, node 12 (or any of the other nodes) may be an Ethernet netwo± adhering to the Ethernet protocol, while node 18 (or any of the other nodes) may be an Fiber Distributed Data Interface (FDDI) network adhering to the FDDI protocol. Furthermore, one or more of the nodes illustrated in 8/30060

FIG. 1 may be of the Asynchronous Transfer Mode (ATM) type, and thus adhere to the ATM protocol.

Referring now to FIG. 2, the switch 20 of FIG. 1 is shown to include a Data Moving Engine (DME) 50, a frame memory 52, and a lookup memory 54. Further, the DME 50 is shown to be conneαed to the ports 24, 26, and 28. In a preferred embodiment, the DME 50 is a custom ASIC designed by Digital Equipment Corporation; it performs unicast transparent bridge forwarding (i.e., switching) at up to 1,000,000 pps, advanced bridge filtering, and core LAN functions at very high speeds. The DME 50 can support up to 48 bridged ports. The lookup memory 54 is used for data structures required by bridging and application code, while the flash memory 52 is used to store the configuration and operational code, and to run the code. When a frame of data arrives on a port, port 24 for example, the DME 50 checks the frame memory 52 and tha lookup memory 54 to insure that the frame contains a valid source address and a valid destination address. Data leaving switch 20 flows from the DME 50 and on to the bus 36 via port 28. Data arriving at switch 20 comes off the bus 36 and flows into the DME 50 via port 28.

Again the DME 50 checks the received data for valid source and destination address in the frame memory 52 and the lookup memory 54, and then forwards the data to the appropriate node via the appropriate port, node 12 via port 24, for example.

The present invention resides in the DME of the switch, DME 50 in switch 20, for example.

A frame of data 60 arriving in the DME 50 from a node, 12 for example, may contain a 8/30060

vaπaoie amount or oytes as illustrated in FIG. 3A. After arriving in the DME 50, a main header 62 is added to the frame of data 60 as shown in FIG. 3B. The main header 62 is used to identify the protocol of the incoming frame of data 60. As now shown in FIG. 3B, the DME 50 adds one or more secondary headers labeled as 64A, 64B, and 64C, respectively, depending on the number of bytes in the frame of data 60 to be transmitted onto the bus 36, and one or more trailers labeled as 66A, 66B, and 66C, respectively. The frame of data 60 of FIG. 3 A is then transmitted on to the bus 36 as a number of smaller bursts labeled as 68 A, 68B , and 68C, respectively , as shown in FIG. 3C. The secondary headers support reassembly of the smaller bursts at the destination or receiver. The transmission then of a number of smaller bursts rather than a larger frame of data simplifies bandwidth allocation and reduces bus access latency on the bus 36.

Referring now to FIG.4, an exemplary secondary 64A header is shown to include a 1 byte format. Bits 0, 1, 2, and 3 contain a MOD_ID 80, at bit 4 CHL field 82, at bit 5 a MORE field 84, at bit 6 a PRI field 86, and at bit 7 a SOP field 88. The MODJtD 80 is a four bit field that provides a unique identification of the module (or node) sourcing the burst onto the bus 36. The MOD_ID 80 is used for reassembling bursts into packets at the destination module(s) (or node). The CHL field 82 defines the burst secondary header length. The MORE field 84 is a bit that may be set (but is not required) to indicate that additional bursts are available for transmission.

The PRI field 86 represents the priority to be used for bus arbitration. A value of 0 indicates a low priority, while a value of 1 indicates a high priority.

Lastly, the SOP field 88 indicates a start of burst. A value of 0 indicates a mid-frame burst /30060

or end of frame, while a value of 1 indicates an SOP bunt or single burst packe_t.

Referring to FIG.5, an exemplary trailer 66A consists of two bytes appended to the end of each smaller burst, 68A for example. Bit 7 of Byte 1 contain an EOP field 90 which refers to end of burst. If the value of the EOP field 90 is 0, then the smaller burst 68 A is not the last burst in the frame 60. If the value of the EOP field 90 is 1, then the smaller burst 68A is the last burst in the frame 60.

Bits 4, 5, and 6 refer to a SEQ_NUM field 92. Bits 0 through 7 of byte 2 include a CRC field 94. In a preferred embodiment, the CRC field 94 is defined by the primitive polynomial X**8 + X**4 + X**3 + X**2 + 1. The reduction of frames to smaller bursts employed by the DME 50 in accordance with the invention may be better understood from the flow chart of FIG 6, where at step 100 a frame of data is received by the DME 50. At step 102,"a primary header is added to the frame. At step 104, if the length of the frame is less than 64 bytes, the process continues to step 106 wherein a secondary header is appended to the frame and a trailer is added to the end of the frame and the burst is transmitted on to the bus 36. If the frame's length is greater than 64 bytes, at step 108 a secondary header is appended to a frame containing 64 bytes of data, a trailer is added to the end of the frame, and the resulting burst transmitted as such on to the bus 36. At step 110, the length of the remaining bytes is determined and if it is greater than 64 bytes, at step 112, a secondary header is appended to a frame having 64 bytes of data, a trailer is added to the end of the frame, and the resulting burst transmitted on to the bus 36. If at step 110 the bytes remaining in the frame is less than 64, at step 106 a secondary header is appended to the frame, a trailer is added to the end of the frame, and the resulting burst is transmitted on to the bus 36.

Having described a preferred embodiment of the invention, it will now become apparent, to one skilled in the an that other embodiments incorporating its concepts may be used. It is felt therefore, that this embodiment should not be limited to the disclosed embodiment, but rather should be limited only by the spirit and scope of the appended claims.

Claims

/30060What is Claimed is:

1. A method of transmitting a plurality of bursts of data comprising the steps of: providing a source node, the source node generating a first frame of data to be transmitted to a bus; providing a netwo┬▒ switch, the switch including; a plurality of ports, a first port connected to the source node; a second port connected to the bus; a data moving engine (DME) connected to the plurality of ports, the DME conditioning the first frame comprising the steps of: receiving the first frame; appending a primary header to the first frame to create a primary frame; detemiining whether the length of the of the first frame is greater than 64 bytes; appending a secondary header and a trailer to the primary frame in response to the step of determining to create a secondary frame; transmitting the secondary frame on to the bus; and repeating the steps appending and transmitting until all bytes in the first frame are transmitted on to the bus.

2. The method of transmitting a plurality of bursts of data according to Claim 1 wherein the first frame is of the Ethernet type. 8/30060

3. The method of transmitting a plurality of bursts of data according to Claim 1 wherein the first frame is of the Fiber Distributed Data Interface (FDDI) type.

4. The method of transmitting a plurality of bursts of data according to Claim 1 wherein the first frame is of the Asynchronous Transfer Mode (ATM) type.

5. The method of transmitting a plurality of bursts of data according to Claim 1 wherein the primary header includes information as to the protocol of the first frame.

6. The method of trarisrmtting a plurality of bursts of data according to Claim 1 wherein the secondary header is one 8-bit byte including: four bits indicating a source identification; one bit indicating a burst header length; one bit indicating whether there additional bytes to transmit in the first frame; one bit indicating a priority to be used for bus arbitration; and one bit indicating a start of burst.

7. The method of trarismitting a plurality of bursts of data according to Claim 1 wherein the trailer is two 8-bit bytes including: one bit containing an EOP field which refers to end of burst; three bits containing a sequence number, and

eight bits indicating a cyclic redundancy check (CRC) value.

8.

The method of transmitting a plurality of bursts of data according to Claim 7 wherein the

CRC value is defined by a primitive polynomial X**8 + X**4 + X**3 + ╧ç**2 + 1.