DATA TRANSMISSION IN A COMMUNICATIONS NETWORK

Abstract

In a method for data transmission in a communications network, data frames containing metadata and payloads are transmitted via point-to-point connections between a respective transmitter and a receiver according to a communications protocol. For each point-to-point connection, a transmission list of the data frames to be transmitted from the transmitter to the receiver is kept. In addition, a priority set of at least three interrupt priorities of pairs with differing significance is provided, and the communications protocol permits each data frame of a transmission list to be assigned an interrupt priority and to interrupt a current transmission of a data frame of the transmission list in order to transfer a data frame that is contained in the transmission list and that has been assigned a higher interrupt priority than the currently transferred data frame.

Claims

1.-12. (canceled)

13. A method for transmitting data in a communication network, said method comprising: transmitting data frames including metadata and useful data via point-to-point connections between a respective transmitter and a receiver according to a communication protocol; transmitting a transmission list of data frames from the transmitter to the receiver is managed for each point-to-point connection; prescribing a priority set of at least three interrupt priorities having significances differing in pairs, the communication protocol allowing each data frame on the transmission list to be allocated an interrupt priority; and interrupting transmission of a data frame on the transmission list having a priority that is to be interrupted in order to transmit a data frame that is on the transmission list and has been a allocated higher interrupt priority on the transmission list than the data frame transmission that is being interrupted.

14. The method of claim 13, wherein the communication network has a point-to-point connection between the transmitter and the receiver, further comprising providing agreement between the transmitter and a receiver of the point-to-point connection on how many different interrupt priorities available for the transmission list of the point-to-point connection.

15. The method of claim 13, further comprising: splitting useful data in a data frame included on the transmission list having an allocated interrupt priority that is lower than the highest interrupt priority available in the priority set into data frame fragments before transmitting the data frame; successively transmitting data frame fragments of the data frame; and ascertaining whether the transmission of the data frame is to be interrupted by checking a respective data frame fragment before sending said data frame fragment of the data frame, and whether the transmission list includes a data frame having a higher allocated interrupt priority than the allocated interrupt priority the data frame of this data frame fragment.

16. The method of claim 15, wherein each data frame fragment of the data frame has a frame length of at least 64 bytes.

17. The method of claim 13, further comprising interrupting the transmission of a data frame if, during the transmission of said data frame, a data frame having a higher allocated interrupt priority than the data frame the data frame being transmitted is incorporated into the transmission list.

18. The method of claim 17, wherein the data frame has a minimum residual number of bytes, and further comprising interrupting the transmission of a data frame on the transmission list only if the number of useful data bytes as yet untransmitted exceeds the prescribed minimum residual number of bytes in the data frame.

19. The method of claim 17, further comprising incorporating a data frame fragment in the transmission list that includes as yet untransmitted useful data of a data frame in the transmission list in response to an interruption in the transmission of a data frame.

20. The method of claim 18, further comprising incorporating a data frame fragment in the transmission list that includes as yet untransmitted useful data of a data frame in the transmission list in response to an interruption in the transmission of a data frame.

21. The method of claim 13, wherein the metadata of each data frame having an allocated interrupt priority that is lower than the highest interrupt priority in the priority set, includes the interrupt priority allocated to the data frame and a frame number allocated to the data frame that characterizes the data frame.

22. The method of claim 15, wherein the metadata of each data frame fragment of a data frame includes the interrupt priority allocated to the data frame, a frame number allocated to the data frame, and the number of useful data bytes of the data frame not yet sent.

23. The method of claim 16, wherein the metadata of each data frame fragment of a data frame includes the interrupt priority allocated to the data frame, a frame number allocated to the data frame, and the number of useful data bytes of the data frame not yet sent.

24. The method of claim 19, wherein the metadata of each data frame fragment of a data frame includes the interrupt priority allocated to the data frame, a frame number allocated to the data frame, and the number of useful data bytes of the data frame not yet sent.

25. The method of claim 21, wherein a multicast address is reserved for the data frame fragments of a respective data frame, the data frame the data frame, and the metadata of each data frame fragment that follows a first data frame fragment, and further comprising including the multicast address in the data frame fragments of a respective data frame as a destination address.

26. The method of claim 22, wherein a receiver receives a data frame split into data frame fragments, and further comprising: using the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number; comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame; and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

27. The method of claim 23, wherein a receiver receives a data frame split into data frame fragments, and further comprising: using a data frame split into data frame fragments uses the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number; comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame, and in that the receiver, and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

28. The method of claim 24, wherein a receiver receives a data frame split into data frame fragments, and further comprising: using the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number; comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame, and in that the receiver; and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

29. The method of claim 25, wherein a receiver receives a data frame split into data frame fragments, and further comprising: using the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number; comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame, and in that the receiver; and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

30. The method of claim 22, wherein a receiver of a data frame split into data frame fragments uses the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number, comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame, and in that the receiver, and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

31. The method of claim 25, wherein a receiver receives a data frame split into data frame fragments, and further comprising: using the metadata of the data frame fragments received by the receiver to ascertain whether it has received all data frame fragments of the data frame by ascertaining the total number of useful data bytes in all data frame fragments it received having metadata including the same frame number, comparing said total number of useful data bytes in the data frame with the total number of received useful data bytes of said data frame, and in that the receiver, and compiling the received data frame fragments having metadata including the same frame number to form a data frame after all data frame fragments of the data frame have been received.

32. The method of claim 13, wherein the communication network is an Ethernet network.

Description

[0026] The properties, features and advantages of this invention that are described above and the way in which they are achieved will become clearer and more distinctly comprehensible in conjunction with the description below of exemplary embodiments that are explained in more detail in conjunction with the drawings, in which:

[0027] FIG. 1 shows a time profile for a transmission of five data frames,

[0028] FIG. 2 shows the structure and sending of an express frame,

[0029] FIG. 3 shows the structure and sending of a non-express frame that is not split into data frame fragments,

[0030] FIG. 4 shows the structure and sending of a non-express frame that is split into three data frame fragments,

[0031] FIG. 5 shows data fields of a preemption tag of a non-express frame,

[0032] FIG. 6 shows the reception of a non-express frame sent without interruption that has not been split into data frame fragments,

[0033] FIG. 7 shows the reception of a non-express frame that has been split into three data frame fragments,

[0034] FIG. 8 shows a flowchart for a first interrupt variant, and

[0035] FIG. 9 shows a flowchart for a second interrupt variant.

[0036] Mutually corresponding parts are provided with the same reference symbols in all figures.

[0037] FIG. 1 shows an example of a time profile for a transmission of five data frames respectively including metadata and useful data via a point-to-point connection from a transmitter to a receiver of a communication network according to a communication protocol. The communication network is in the form of an Ethernet network. Accordingly, the communication protocol is an Ethernet protocol and the data frames are Ethernet frames. Each data frame on a transmission list of the transmitter is allocated an interrupt priority of a prescribed priority set of at least three interrupt priorities having significances differing in pairs. Data frames that have the highest allocated interrupt priority included in the priority set are referred to as express frames. All other data frames are referred to as non-express frames.

[0038] The transmission of a non-express frame on the transmission list is interrupted according to the communication protocol if, during the transmission, a data frame is incorporated into the transmission list that has a higher allocated interrupt priority than the presently transmitted non-express frame, provided that an as yet untransmitted number of useful data bytes of the non-express frame exceeds a minimum residual number of bytes, which is prescribed as 44 bytes, for example.

[0039] To this end, a first interrupt variant provides for the useful data of each non-express frame that is included in a transmission list and that includes more useful data bytes than a prescribed minimum number of bytes, for example more than 88 bytes, to be split, before the transmission of the non-express frame, over data frame fragments that are transmitted in succession. Before the sending of a data frame fragment of the non-express frame, a respective check is performed to determine whether the transmission list includes a data frame that has a higher allocated interrupt priority than the non-express frame of this data frame fragment. Should the transmission list include such a data frame, the transmission of the data frame fragment is deferred and this data frame starts to be transmitted (in this regard see FIG. 8).

[0040] Alternatively, a second interrupt variant provides for the non-express frames on the transmission list to be split into data frame fragments not before they are actually transmitted but rather only if, during the transmission of a non-express frame, a data frame that has a higher allocated interrupt priority than the presently transmitted non-express frame is incorporated into the transmission list, provided that the as yet untransmitted number of useful data bytes of the non-express frame exceeds the minimum residual number of bytes. In this case, the transmission of the non-express frame is interrupted, a data frame fragment that includes the as yet untransmitted useful data of the non-express frame is formed and incorporated into the transmission list, and the transmission of the data frame having the higher interrupt priority is started.

[0041] By contrast, express frames are always transmitted without an interrupt.

[0042] In FIG. 1, EF_1 and EF_2 denote a first express frame and a second express frame, NEF2 Frag_1, NEF2 Frag_2 and NEF2 Frag_3 denote three different data frame fragments of a first non-express frame that has the third highest allocated interrupt priority included in the priority set, NEF1 Frag_1, NEF1 Frag_2 and NEF1 Frag_3 denote three different data frame fragments of a second non-express frame that has the second highest allocated interrupt priority included in the priority set, and NEF1 denotes a third non-express frame that likewise has the second highest allocated interrupt priority included in the priority set.

[0043] First of all, a first data frame fragment NEF2 Frag_1 of the first non-express frame is transmitted. The transmission of the first non-express frame is interrupted to transmit the second non-express frame. Therefore, a first data frame fragment NEF1 Frag_1 of the second non-express frame is subsequently transmitted. The transmission of the second non-express frame is subsequently interrupted to transmit the first express frame EF_1. After the transmission of the first express frame EF_1, the transmission of the second non-express frame that was interrupted thereby is continued by virtue of a second data frame fragment NEF1 Frag_2 of the second non-express frame being transmitted. The transmission of the second non-express frame is then interrupted again to transmit the second express frame EF_2. After the transmission of the second express frame EF_2, the transmission of the second non-express frame that was interrupted thereby is continued by virtue of the third and last data frame fragment NEF1 Frag_3 of the second non-express frame being transmitted. Subsequently, the interrupted transmission of the first non-express frame is continued by virtue of a second data frame fragment NEF2 Frag_2 of the first non-express frame being transmitted. The transmission of the first non-express frame is then interrupted again to transmit the third non-express frame NEF1. The third non-express frame NEF1 is not split into data frame fragments and transmitted without interruption, for example because, in the event of the use of the first interrupt variant, it does not include more useful data bytes than the prescribed minimum number of bytes or because, in the event of the use of the second interrupt variant, no express frame is incorporated into the transmission list during the transmission of the third non-express frame NEF1. After the transmission of the third non-express frame NEF1, the transmission of the first non-express frame is continued and is completed by the transmission of the third and last data frame fragment NEF2 Frag_3 of the first non-express frame.

[0044] The communication protocol differs from known Ethernet protocols in that it allows multi-level preemption of the transmission of data frames. This multi-level preemption is understood to mean the above-described interruption of the transmission of a non-express frame for the purpose of transmitting an express frame or a non-express frame that has a higher allocated interrupt priority than the presently transmitted non-express frame. In FIG. 1, it has been assumed that both the transmitter and the receiver of the point-to-point connection support this multi-level preemption. Whether and to what extent the transmitter and the receiver of a point-to-point connection support multi-level preemption is clarified between the transmitter and the receiver before the transmission of the data frames, for example by means of a negotiation protocol such as the link layer discover protocol (LLDP). Should this reveal that the transmitter or the receiver does not support multi-level preemption, data frames from the transmitter are transmitted to the receiver without multi-level preemption. In the event of both the transmitter and the receiver supporting multi-level preemption, the transmitter and the receiver, before the transmission of data frames, clarify particularly among one another how many different interrupt priorities are made available for the multi-level preemption and into how many data frame fragments a non-express frame is broken down at most for the multi-level preemption.

[0045] FIGS. 2 to 4 schematically show the structure and sending of data frames. In this case, FIG. 2 shows the structure and sending of an express frame EF, FIG. 3 shows the structure and sending of a non-express frame NEF1 that is not split into data frame fragments, and FIG. 4 shows the structure and sending of a non-express frame NEF2 that is split into three data frame fragments.

[0046] According to FIG. 2, an express frame EF has a header for metadata that precedes the useful data P included in the express frame EF. The metadata included in the header comprise a receiver address DA, which is an MAC address (=Media Access Control address) of the receiver, a transmitter address SA, which is an MAC address of the transmitter, an optional VLAN tag VLAN and an EtherType entry EtherType to denote the protocol needed for interpreting the useful data P. The express frame EF is augmented with a checksum FCS by a transmission module 1 at a transmission port of the transmitter, which is connected to a receiver port of the receiver, to identify transmission errors and is sent to the receiver port of the receiver. The checksum FCS is formed from the metadata and useful data P of the express frame EF in a known manner, for example as a CRC checksum (CRC=Cyclic Redundancy Check), and inserted after the useful data P as a trailer of the sent express frame EF.

[0047] FIG. 3 shows the structure and sending of a non-express frame NEF1 that is not split into data frame fragments and whose transmission is not interrupted. Like the express frame EF depicted in FIG. 2, the non-express frame NEF1 has a header for metadata that comprise a receiver address DA, a transmitter address SA, an optional VLAN tag VLAN and an EtherType entry EtherType. In order to distinguish a non-express frame NEF1 from an express frame EF and describe its multi-level preemption properties, its header has metadata added to it by a preemption module 3, said metadata comprising an EtherType supplementary entry IET_EtherType and a preemption tag IET_Tag. The EtherType supplementary entry IET_EtherType distinguishes the non-express frame NEF1 from an express frame EF. The preemption tag IET_Tag includes information relevant to the multi-level preemption, which information is described in more detail below with reference to FIG. 5. After the addition of the EtherType supplementary entry IET_EtherType and of the preemption tag IET_Tag, the non-express frame NEF1 is, like an express frame EF, augmented with a checksum FCS by a transmission module 1 at a transmission port of the transmitter, which is connected to a receiver port of the receiver, to identify transmission errors and is sent to the receiver port of the receiver.

[0048] FIG. 4 shows the structure and sending of a non-express frame NEF2 whose useful data P are split over three data frame fragments Frag1, Frag2, Frag3 before the transmission of said non-express frame from the transmitter to the receiver according to the first interrupt variant described above. Like the express frame EF depicted in FIG. 2, the non-express frame NEF2 has a header for metadata that comprise a receiver address DA, a transmitter address SA, an optional VLAN tag VLAN and an EtherType entry EtherType. A preemption module 3 forms three data frame fragments Frag1, Frag2, Frag3 from the express frame NEF2. A first data frame fragment Frag1 includes a first useful data component P1 of the useful data P of the non-express frame NEF2 and has a header that, like the header of the non-express frame NEF1 depicted in FIG. 3, comprises as metadata the receiver address DA, the transmitter address SA, the optional VLAN tag VLAN and the EtherType entry EtherType and also an EtherType supplementary entry IET_EtherType and a preemption tag IET_Tag. The two further data frame fragments Frag2, Frag3 respectively include a further useful data component P2, P3 of the useful data P of the express frame NEF2 and respectively have a header that includes as metadata a multicast address DA Peer reserved for data frame fragments, an MAC port address SA_IF specific to the respective transmission port of the transmitter and also an EtherType supplementary entry IET_EtherType and a preemption tag IET_Tag. A transmission module 1 at the transmission port augments each data frame fragment Frag1, Frag2, Frag3 formed by the preemption module 3 with a checksum FCS to identify transmission errors and sends said data frame fragments to a receiver port of the receiver.

[0049] FIG. 5 shows, by way of example, data fields of a preemption tag IET_Tag of a non-express frame NEF1, NEF2 that are split over three bytes, each byte of the preemption tag IET_Tag being depicted in one row in FIG. 5.

[0050] A first data field D1 that comprises eight bits and a second data field D2 that comprises three bits include a number of as yet unsent useful data bytes of the non-express frame NEF1, NEF2. The total number eleven of the bits of these data fields D1, D2 in this exemplary embodiment matches the maximum number 1500 of useful data bytes that are usually provided for an Ethernet frame. On the basis of the number of as yet unsent useful data bytes that is included in the data fields D1, D2 and the number of already received useful data bytes of a non-express frame NEF1, NEF2, the receiver can identify whether it has already received all useful data P of the non-express frame NEF1, NEF2.

[0051] A third data field D3 that comprises three bits includes a frame number that is allocated to the non-express frame NEF1, NEF2 and on the basis of which different non-express frames NEF1, NEF2 on the same transmission list are distinguished. The preemption tag IET-Tag depicted in FIG. 5 can thus assign eight different frame numbers, and hence eight non-express frames NEF1, NEF2 can be distinguished from one another. If a non-express frame NEF1, NEF2 is split into multiple data frame fragments Frag1, Frag2, Frag3 (see FIG. 4), all these data frame fragments Frag1, Frag2, Frag3 are provided with the same frame number, which means that the receiver can identify data frame fragments Frag1, Frag2, Frag3 belonging to the same non-express frame NEF1, NEF2.

[0052] Further fourth data fields D4 that respectively comprise one bit include the interrupt priority allocated to the non-express frame NEF1, NEF2. In this case, the number of fourth data fields D4 needed therefor is dependent on the number of interrupt priorities included in the priority set. FIG. 5 depicts ten fourth data fields D4. If the priority set includes only three different interrupt priorities, for example, however, then denoting the interrupt priority of a non-express frame NEF1, NEF2 requires only one fourth data field D4, since the priority set includes only two different interrupt priorities that can be allocated to a non-express frame NEF1, NEF2 (since the highest interrupt priority is reserved for express frames E4 in each case, of course). If, by contrast, the priority set includes five different interrupt priorities, for example, then denoting the interrupt priority of a non-express frame NEF1, NEF2 requires two fourth data fields D4. A similar situation applies to higher numbers in the priority set of included interrupt priorities.

[0053] FIGS. 6 and 7 respectively show the reception of a non-express frame NEF1, NEF2 at a receiver port of a receiver of the non-express frame NEF1, NEF2.

[0054] FIG. 6 shows the reception of a non-express frame NEF1 that has been sent without interruption according to FIG. 3 and that has not been split into data frame fragments. The non-express frame NEF1 is received by a receiver module 5 at the reception port. The receiver module 5 uses the checksum FCS to check whether the non-express frame NEF1 has been received correctly. If the result of this check turns out to be positive, the receiver module 5 removes the checksum FCS and transfers the non-express frame NEF1 to a defragmentation module 7. The defragmentation module 7 uses the metadata in the header of non-express frame NEF1 to identify that the non-express frame NEF1 is not split into data frame fragments and has been received completely, and removes the EtherType supplementary entry IET_EtherType and the preemption tag IET_Tag from the header.

[0055] FIG. 7 shows the reception of a non-express frame NEF2 that has been sent according to FIG. 4 and that has been split into three data frame fragments Frag1, Frag2, Frag3. The three data frame fragments Frag1, Frag2, Frag3, which respectively include a checksum FCS, are received by a receiver module 5 at the reception port in succession. The receiver module 5 checks for each data frame fragment Frag1, Frag2, Frag3, using the checksum FCS thereof, whether the data frame fragment Frag1, Frag2, Frag3 has been received correctly. If the result of this check turns out to be positive, the receiver module 5 removes the respective checksum FCS and transfers the respective data frame fragment Frag1, Frag2, Frag3 to a defragmentation module 7. The defragmentation module 7 uses the metadata in the headers of the data frame fragments Frag1, Frag2, Frag3 to identify that the non-express frame NEF2 is split into data frame fragments Frag1, Frag2, Frag3 and when all three data frame fragments Frag1, Frag2, Frag3 have been received completely. After all three data frame fragments Frag1, Frag2, Frag3 have been received completely, the defragmentation module 7 recompiles the data frame fragments Frag1, Frag2, Frag3 to form the original non-express frame NEF2, removing the EtherType supplementary entry IET_EtherType and the preemption tag IET_Tag from the header.

[0056] According to FIGS. 2 to 4, 6 and 7, the transmission modules 1 merely add a respective checksum FCS to data frames they send and the receiver modules 5 respectively remove merely the checksum FCS from the data frames they receive, the addition and removal of a checksum FCS respectively being performed in the same manner for an express frame, a non-express frame that has not been split into data frame fragments and a data frame fragment of a non-express frame. Therefore, the processing of express frames, non-express frames and data frame fragments does not require any special or additional transmission modules 1 or receiver modules 5 in comparison with the use of a conventional Ethernet protocol without multi-level preemption.

[0057] The EtherType supplementary entry IET_EtherType and the preemption tag IET_Tag are used merely to realize the transmission interrupts with multi-level preemption and are removed again after the reception of non-express frames, so that these frames, after their reception, can be processed in their original form by higher layers of the 051 model (=Open Systems Interconnection model).

[0058] FIG. 8 shows a flowchart for the first interrupt variant already mentioned above for a transmission list for the case in which the priority set includes five different interrupt priorities. Subsequently, the highest of these interrupt priorities is referred to as interrupt priority 0, the second highest is referred to as interrupt priority 1, the third highest is referred to as interrupt priority 2, the fourth highest is referred to as interrupt priority 3 and the fifth highest is referred to as interrupt priority 4. Each express frame is thus allocated the interrupt priority 0, while each non-express frame is allocated one of the interrupt priorities 1 to 4. The flowchart depicted in FIG. 8 includes method steps S1 to S9.4, which are described below.

[0059] S1: A data frame that is subsequently referred to as the data frame under consideration and has a highest interrupt priority among all data frames presently included in the transmission list and not processed is taken from the transmission list.

[0060] S2: A check is performed to determine whether the data frame under consideration is a non-express frame, for example by checking whether it has an EtherType supplementary entry IET_EtherType. If this is not the case, the data frame under consideration is an express frame and S3 is carried out; otherwise, the data frame under consideration is a non-express frame and S4 is carried out.

[0061] S3: The data frame under consideration, which in this case is an express frame, is sent (see FIG. 2 and the description thereof). Subsequently, S1 is carried out again.

[0062] S4: A check is performed, for example by means of an LLDP, to determine whether both the transmitter and the receiver support multi-level preemption. If this is not the case, S5 is carried out; otherwise, S6.1 is carried out.

[0063] S5: The data frame under consideration, which in this case is a non-express frame, is sent without interruption (see FIG. 3 and the description thereof). Subsequently, S1 is carried out again.

[0064] S6.X for X=1, 2, 3, 4: A check is performed to determine whether the transmission list includes a data frame fragment Frag2, Frag3 yet to be sent having the interrupt priority X. If this is the case, S7.X is carried out; otherwise, S8.X is carried out.

[0065] S7.X for X=1, 2, 3, 4: The data frame fragment Frag2, Frag3 yet to be sent ascertained in S6.X, having the interrupt priority X, is sent. Subsequently, S1 is carried out.

[0066] S8.X for X=1, 2, 3, 4: A check is performed to determine whether the transmission list includes a non-express frame having the interrupt priority X. If this is not the case, then the method step S6.X+1 is carried out in the cases X=1, 2, 3 and method step S1 is carried out in the case X=4; otherwise, S9.X is carried out.

[0067] S9.X for X=1, 2, 3, 4: A non-express frame ascertained in S8.X, having the interrupt priority X, is split into data frame fragments Frag1, Frag2, Frag3 as described above if need be, i.e. if this non-express frame contains more useful data bytes than the prescribed minimum number of bytes. If the non-express frame is split into data frame fragments Frag1 Frag2, Frag3, the first data frame fragment Frag1 is sent (see FIG. 4 and the description thereof) and the further data frame fragments Frag2, Frag3 are incorporated into the transmission list. Otherwise, i.e. if the non-express frame is not split into data frame fragments, the non-express frame is sent without interruption (see FIG. 3 and the description thereof). Subsequently, S1 is carried out.

[0068] FIG. 9 shows a flowchart for the second interrupt variant already mentioned above for a transmission list for the case in which the priority set includes five different interrupt priorities. As in the description of FIG. 8, the highest of these interrupt priorities is referred to as interrupt priority 0, the second highest is referred to as interrupt priority 1, the third highest is referred to as interrupt priority 2, the fourth highest is referred to as interrupt priority 3, and the fifth highest is referred to as interrupt priority 4. Each express frame is thus allocated the interrupt priority 0, while each non-express frame is allocated one of the interrupt priorities 1 to 4. The flowchart depicted in FIG. 9 includes method steps S1 and S10 to S20, which are described below.

[0069] S1: A data frame that is subsequently referred to as the data frame under consideration and has a highest interrupt priority among all data frames presently included in the transmission list and not processed is taken from the transmission list.

[0070] S10: A check is performed, for example by means of an LLDP, to determine whether both the transmitter and the receiver support multi-level preemption. If this is not the case, S11 is carried out; otherwise, S12 is carried out.

[0071] S11: The data frame under consideration is sent. Subsequently, S1 is carried out.

[0072] S12: A check is performed to determine whether the data frame under consideration is a non-express frame, for example by checking whether it has an EtherType supplementary entry IET_EtherType. If this is not the case, the data frame under consideration is an express frame and S13 is carried out; otherwise, the data frame under consideration is a non-express frame and S16.1 is carried out.

[0073] S13: A check is performed to determine whether a non-express frame is presently being sent. If this is not the case, S14 is carried out; otherwise, S15 is carried out.

[0074] S14: The data frame under consideration, which in this case is an express frame, is transmitted completely. Subsequently, S1 is carried out.

[0075] S15: If the as yet untransmitted number of useful data bytes of the presently sent non-express frame exceeds the minimum residual number of bytes, the transmission of the presently sent non-express frame is interrupted and a data frame fragment that includes the as yet untransmitted useful data of the non-express frame is formed and incorporated into the transmission list. Otherwise, the transmission of the presently sent non-express frame is completed. In any case, S14 is subsequently performed.

[0076] S16.X for X=1, 2, 3, 4: A check is performed to determine whether a non-express frame having the interrupt priority X is currently being sent. If this is the case, S1 is carried out; otherwise, method step S17.X is carried out in the cases X=1, 2, 3 and method step S19 is carried out in the case X=4.

[0077] S17.X for X=1, 2, 3: A check is performed to determine whether the data frame under consideration has the interrupt priority X. If this is not the case, S16.X+1 is carried out; otherwise, S18 is carried out.

[0078] S18: A check is performed to determine whether a non-express frame is currently being sent. If this is not the case, S19 is carried out; otherwise, S20 is carried out.

[0079] S19: The data frame under consideration starts to be transmitted. Subsequently, S1 is carried out.

[0080] S20: If the as yet untransmitted number of useful data bytes of the presently sent non-express frame exceeds the minimum residual number of bytes, the transmission of the presently sent non-express frame is interrupted and a data frame fragment that includes the as yet untransmitted useful data of the non-express frame is formed and incorporated into the transmission list. Otherwise, the transmission of the presently sent non-express frame is completed. In any case, S19 is subsequently performed.

[0081] Although the invention has been illustrated and described in more detail by preferred exemplary embodiments, the invention is not limited by the examples disclosed and other variations can be derived therefrom by a person skilled in the art without departing from the scope of protection of the invention.

LIST OF REFERENCE SYMBOLS

[0082] 1 Transmission module [0083] 3 Preemption module [0084] 5 Receiver module [0085] 7 Defragmentation module [0086] D1 to D4 Data field [0087] DA Receiver address [0088] DA_Peer Multicast address [0089] EF, EF_1, EF_2 Express frame [0090] EtherType EtherType entry [0091] FCS Checksum [0092] Frag1, Frag2, Frag3 Data frame fragment [0093] IET-EtherType EtherType supplementary entry [0094] IET_Tag Preemption tag [0095] NEF1, NEF2 Non-express frame [0096] NEF1 Frag_1, NEF1 Frag_2, NEF1 Frag_3 Data frame fragment [0097] NEF2 Frag_1, NEF2 Frag_2, NEF2 Frag_3 Data frame fragment [0098] P Useful data [0099] P1, P2, P3 Useful data component [0100] S1 to S20 Method step [0101] SA Transmitter address [0102] SA_IF MAC port address [0103] VLAN VLAN tag