WO2007042343A1

WO2007042343A1 - Method for transferring data by means of ethernet frames and network element for processing ethernet frames.

Info

Publication number: WO2007042343A1
Application number: PCT/EP2006/065577
Authority: WO
Inventors: Andreas Fricke
Original assignee: Nokia Siemens Networks Gmbh & Co. Kg
Priority date: 2005-10-13
Filing date: 2006-08-23
Publication date: 2007-04-19
Also published as: DE102005048966A1

Abstract

The invention relates to a method for transferring data by means of ethernet frames and to a network element for processing ethernet frames. Said frames comprise a preamble and are transferred from a first a network element to a second network element. At least one first and one second transfer path are provided at least partially between both network elements. The data is transferred via at least the first transfer path. A first and a second byte is provided in the ethernet preamble, wherein in the first byte, information of the Kl-byte and in the second byte, information is introduced according to the K2-byte of the SDH-norm ITU-T G.841. A disturbance of the first transfer path is detected by one of the two transfer paths and the data is transferred via the second network element, whereby this is signalised, by means of the KI-byte information and/or K2-byte information which is contained in the first and seconds bytes, to the other network element.

Description

description

Method for transmitting data by means of Ethernet frame and network element for processing Ethernet frames.

The invention relates to a method according to the preamble of patent claim 1 and a network element for processing Ethernet frames.

For the transmission of data the so-called Ethernet is often used. Ethernet is a frame-based or frame-based transmission technology for the transmission of data. Ethernet includes a connectionless protocol for data transmission, which is widely used because of its simplicity.

The object of the present invention is to improve the transmission of data by means of Ethernet frames.

This object is achieved by a method having the features of patent claim 1 and a network element having the features of patent claim 6.

According to the invention, it is proposed that data is transmitted via at least one transmission path, and that, in the event of a disturbance of this transmission path, a second undisturbed transmission path is used. The signaling of the fault or the changeover takes place by means of information which is transmitted in the preamble of the Ethernet frame. Here, the information is exchanged using the Kl and K2 byte functionality according to the SDH standard ITU-T G.841. For this the Kl and the K2 byte are transferred into two bytes of the preamble. The meaning, coding, functionality and use of the K bytes in the network elements for processing Ethernet frames or Ethernet frames is identical to the SDH standard. This has the advantage that a fast equivalent circuit in case of failure of a connection path or Ether- net segment is given for the data transmission, if another connection path is available. Switching is very fast in this case and can be in the microsecond range. This leads to an increase in the attractiveness of Ethernet.

Advantageous embodiments of the invention are specified in the dependent claims.

An embodiment of the invention will be explained in more detail below with reference to the drawing. Showing:

FIG. 1 shows a preamble of an Ethernet frame,

Figure 2 is a circuit diagram for the inventive part of a network element.

FIG. 1 shows a preamble of an Ethernet frame or Ethernet frame. This preamble has 8 bytes. The first byte SOP is the so-called start-off packet. The next 2 bytes of OAM are called Operation, Administration and Maintanance information. According to the invention, in the two following two bytes, the information according to the K bytes of the SDH standard is inserted according to ITU-T G.841, ie the information of the Kl and K2 bytes of the abovementioned standard or the Kl and K2 bytes , These two bytes are therefore labeled Kl and K2. The following two bytes, labeled Logical-Phy ID, contain logical-physical-layer information. The last byte CRC contains a code word backup sequence / cyclic redundancy check. This is followed by the Ethernet MAC Frame.

Figure 2 shows a section of a circuit diagram for a network element, with a first Ubertragungsweg respectively a ner working line WLl, which is connected to a first Ethernet Phy device respectively Ethernet physical layer device ETH Phyl. In an analogous manner, further transmissions tion paths / working line WL2, ..., WLN are each connected to an Ethernet Phy block ETH Phy2, ..., ETH PhyN. Furthermore, a replacement transmission path or Protection Line WLP is provided, which is connected to an Ethernet Phy module ETH PhyP. The Ethernet Phy modules ETH Phyl, ..., ETH PhyN, ETH PhyP are each connected to a switching device SE. On the other hand, this is connected in each case to an Ethernet MAC module or Ethernet media access control module ETH MAC1,..., ETH MACN, ETH MACP, with one Ethernet Phy component each being connected to an Ethernet MAC module by the switching device SE is. In the example of the first Ethernet Phy module ETH Phyl with the first Ethernet MAC module ETH MACl, ETH Phy2 with ETH MAC2 etc. This means that an Ethernet frame that comes from a working line is sent to the Ethernet via the Ethernet Phy module and the switching device -MAC unit passed. This applies analogously to frameworks of the Protection Line.

If a working line has a fault, this is detected by a network element and a corresponding information K1 / K2 encoded by the Ethernet Phy block is inserted in the preamble of the Ethernet frame and transmitted to the adjacent network element. There, the K1 / K2 byte of the preamble is evaluated in the Ethernet Phy block. Switching in the network element (the first and the second network element) is performed such that a switching command is sent from the corresponding Ethernet Phy component to the switching device SE, so that it switches from the disturbed or failed working line to the protection line / Ersatzübertragungsweg. The data traffic of the Protection-Line is rejected here.

The processing of the K1 / K2 bytes can also be done centrally by means of a K-byte processing device KV, as shown in FIG. In this case, a detected or a notified (by means of K1 / K2 byte) disturbance is given to the K-byte processing function, which on the one hand gives a switching command to the switching device SE in the network element, so that from a faulty working line is switched to a protection line and on the other hand causes the transmission of the switching information using K1 / K2 bytes on the Ersatzübertragungsweg.

According to the preamble of an Ethernet frame is modified as shown in Figure 1. Deviating from the standard, two bytes are provided which carry information of the K1 and K2 bytes whose meaning, coding, functionality and use of the K bytes are identical to the SDH standard as described in ITU-T G.841 are. Thus, e.g. unidirectional protection, bidirectional protection, 1: N protection or ring protection. Such a preamble is generated or terminated and processed in the Ethernet Phy block or in the Ethernet physical layer interface. The processing can also be carried out centrally by a K-byte processing function, which controls a corresponding switching device, so that a trouble-free connection path for the Ethernet data transmission is available. The Ethernet Phy device becomes a

Standard preamble to the media access control device or to the MAC module. Using the preamble for the control information does not increase the data rate of the Ethernet.

According to FIG. 2, the traffic is forwarded by the Phy components to the MAC components in the event of a fault. The modified preamble is filtered out by the Phy blocks and replaced by standard preambles, which are given to the MAC blocks via the switching device so that they do not require any adaptation.

A fault can be detected or triggered by - forced switching to the protection path - signaling from the opposite side by the K-byte protocol (K1, K2) a loss-of-bit synchronization of the physical layer

Interfaces z. Eg no signal

a loss-of-byte synchronization of the physical layer

Interfaces - a link-down message

a rate of erroneous OAM (Operation Administration Maintenance Praamble), detected by checking the OAM CRC, ie the check sum of the OAM frames through the physical layer interface function, by means of a physical layer link quality Supervision, d. H. if the signal quality is not sufficient, is reduced to a lower link speed z. B. switched from 1 GBit per second to 100 MBit per second, but a replacement link used - if an autonegotiation went wrong

- If the rate of faulty Ethernet frames exceeds a threshold, which is done by checking the MAC check sum CRC, which can be realized in the MAC function.

In the event of an error, the network element detecting the error signals the opposite network element with the aid of the K 1, K 2 byte information that the protection link / replacement transmission path or second transmission path is switched over. Then both devices switch to the protection line at the end of the broken link. This is shown in Figure 2 rudimentary by dashed switch in the switching device SE, wherein the switch of the protection line occupies the dashed line position and the switch of the disturbed working line / the disturbed Ubertragungsweges also occupies the dashed line position. This means that all traffic is transmitted bidirectionally via the Protection Line. Since the switching takes place before the MAC module, all Ethernet frames or Ethernet frames destroyed by the switching process are filtered out and paid. The actual performance including all switching operations can be monitored. In order to achieve switching times that are comparable to the SDH standard or SONET standard, which is usually less than 50 ms, it must be ensured that at least one preamble with K bytes is transmitted per unit of time. This is generally not the case with Ethernet. If no data / Ethernet traffic is available, no Ethernet frames are generated and transmitted, which means that no preamble is transmitted. For this reason it is proposed to transfer so-called "dummy preambles", ie, preambles without the following MAC frame or without further useful content, if no Ethernet frames are transmitted in a defined period of time.These dummy preambles should be transmitted by the Ethernet Phy Block or be completely terminated and filtered out by the physical layer interface function.

The following table shows the duration of an Ethernet frame for Ethernet data rates from 10 Mb / s to 10 Gb / s and frame lengths of 1600 bytes and 9600 bytes (jumbo frame).

By comparison, SDH frames have a duration of 125 μs. A 10 Mb / s Ethernet transmission could thus not achieve the switching times according to the SDH standard.

A persistence check should not propagate K byte changes until after three consecutive frames as a change. A defect which is signaled via a K byte protocol can thus be recognized at 100 Mb / s or higher transmission rate after 300 μs. This detection time is slightly faster than the SDH procedure. If the distance between two frames is greater than 150 μs, a phantom Preamble be used. This achieves a reaction rate comparable to that of SDH, although this is 450 μs (or faster), even if no user data is transferred temporarily.

In summary, a method for protecting Ethernet transmission paths using protection circuit functions in accordance with the SDH standard results. Increased security of Ethernet connections with downtime of <50 ms in case of failure is achieved. This makes it possible to achieve a higher acceptance and attractiveness of Ethernet.

Claims

claims

1. A method for transmitting data by means of Ethernet frames, which have a preamble, from a first network element to a second network element, wherein between the two

At least partially at least a first and a second Ubertragungsweg are provided network elements, characterized in that the data via at least the first Übertragungsweg u- are transmitted, that in the Ethernet preamble, a first and a second byte are provided in the first byte information of the Kl-byte and in the second byte information in accordance with the K2 byte of the SDH standard ITU-T G.841 be inserted, that a disturbance of the first Ubertragungsweges is detected by one of the two network elements and the data is transmitted over the second Ubertragungsweg, wherein this is signaled to the other network element by means of the Kl bytes and / or K2 byte information contained in the first and second bytes.

2. The method according to claim 1, characterized in that the first and the second byte are each generated and evaluated in an Ethernet Phy block.

3. The method according to claim 2, characterized in that a changeover from the first to the second transmission path is carried out by means of a switching device which is arranged between the Ethernet Phy and an Ethernet MAC module.

4. The method of claim 1, 2 or 3, characterized in that at least Ethernet preambles are generated and transmitted cyclically.

5. The method according to any one of the preceding claims, characterized in that a modified K-byte must occur in 3 consecutive frames to be passed as a change of the K-byte.

6. Network element for processing Ethernet frames, comprising means for carrying out the method according to one of the preceding claims.

7. Network element according to claim 6, characterized in that in each case a connection path is connected to an Ethernet module which has at least one Ethernet Phy and one Ethernet MAC component, a switching device being arranged between them which is controllable in such a way is switched on at least one remaining connection path in case of failure of a connection path.