JP4350610B2 - Data processing method and data processing apparatus - Google Patents

Description

  The present invention relates to a data processing method related to various devices such as routers in an information communication network system, and in particular, a data processing method for mapping a data stream such as Ethernet (registered trademark) to a synchronous frame such as SONET / SDH. And a data processing apparatus.

In network devices that perform packet switching processing such as routers, the packet processing engine has been implemented exclusively by ASIC (Application Specific Integrated Circuit) so far, but its development requires a large investment.
On the other hand, in recent years, a product group called a network processor has attracted attention. The network processor is a processor specialized for processing network traffic, that is, packets. The network processor is a programmable device, and can be versatile by customizing functions by software.
In addition, efforts to standardize physical interfaces and software interface (Application Program Interface: API) related to network processors are being promoted by corporate consortiums such as the Network Processing Forum (NPF).
If this network processor can be applied to router development as a general-purpose component, significant cost reductions and shortening of the development period can be expected.

However, even if network processors become widespread and the interfaces around them become standardized, problems remain in configuring a router. It is a matter of switch fabric, especially its physical interface. The switch fabric is a chipset consisting of a switch core and IF devices for all manufacturers.
As a role of the IF device, for example, when the switch fabric is a cell-based system, it is natural that the packet is divided into cells at the front stage of the switch core. However, this IF device adds information used for controlling a route inside the switch to a cell or packet as a header.
Depending on the switch chip manufacturer, this IF device may also function as a traffic manager. That is, the header information added by this IF device depends on the internal structure of the switch fabric. In other words, it is unique to the manufacturer, and should be said to be a key point for the switch chip manufacturer to differentiate from other companies' products.
That is, the IF device converts a packet received from the network processor (via a standardized interface) into a format unique to the manufacturer. Therefore, in a router having a general chassis-based form factor, the interface on the backplane that connects the line card and the switch fabric has a unique specification of the switch chip manufacturer.
This is a factor that hinders the multi-source use of parts in router development, and hence the cost reduction of the router.

The physical speed of this backplane is generally a 2.5 to 3.125 Gbps serial link. 8B / 1OB encoding is used for word synchronization (and clock signal embedding), and the effective speed is 2 to 2.5. Gbps. In addition, due to the added header, fixed-length cell, and the like, the substantial data transfer rate is further reduced.
For data traffic, the required backplane physical speed varies depending on the manufacturer's specifications, but it is usually 1.5 to 2 times, sometimes worse than 3 times.

As for the physical speed of this backplane, chip manufacturers who developed 6.25Gbps CMOS (Complementary Metal Oxide Semiconductor) drivers also appeared in 2004, and it is thought that it will further increase to 11Gbps in the future. .
Router switch fabric chips have not only performance such as scalability, but also the ability to immediately follow such technological advances.

  Therefore, we want to apply more open and withered technology to the backplane interface. For example, if a SONET / SDH (ANSI T1. 105 / ITU-T G. 707) switch chip that can be procured as a general-purpose component can be applied to the router switch fabric, the interface of the router backplane can be It can be a SONET / SDH interface such as -x / STS-x.

In other words, if the configuration is such that the line cards are connected in full mesh via SONET / SDH switch chip, and the packet is mapped to the SONET / SDH tributary connected to the destination port, the same function as the packet switch is achieved. You can have it. This makes the backplane interface a standardized interface that does not contain any proprietary format.
The SONET / SDH switch chip is a time lot interchange in STS-1 units, and one chip with a switch capacity of 160 to 340Gbps is available.

  The requirement necessary to realize such a router to which the SONET / SDH switch chip is applied results in the mapping of packet-based traffic to the SONET / SDH tributary. From a different point of view, this can be seen as forwarding packet-based traffic over SONET / SDH.

A typical example of packet-based traffic is Ethernet (registered trademark), that is, LAN (local area network) standardized by IEEE802.3, one of the most popular standards at present.
In Ethernet (registered trademark) and other signal transmission standards, transmission line coding is generally performed. For example, an 8B / 10B code is used for serial Gigabit Ethernet (registered trademark), and a 64B / 66B code is used for 10 Gigabit Ethernet (registered trademark). Therefore, packet-based traffic transmitted serially is generally an encoded digital data signal.

Recently, Ethernet (registered trademark) is also attracting attention as a transport technology in MAN (metro area network), and the demand for wide area Ethernet (registered trademark) service (TLS: Transparent LAN Service) is growing significantly. .
In addition, 10-Gigabit Ethernet (registered trademark) (IEEE 802.3ae) is being extended to WAN. The 10-Gigabit Ethernet (registered trademark) interface is also beginning to appear on the market as a practical product, and is expected to be cheaper in the future.

  On the other hand, SONET / SDH is still supported by carriers (first-class telecommunications carriers) because of its ability to provide various granularity lines, rich management functions, and existing facilities. Have gained. From the carrier's standpoint, it is desirable to be able to transfer Ethernet (registered trademark) (especially 10 Gigabit Ethernet (registered trademark)) traffic, which is expected to grow in the future, using SONET / SDH equipment.

The Ethernet (registered trademark) standard has been speeded up by a factor of 10 because it has progressed in line with the maturity of component technology rather than taking into account the requirements of a network device. As a result, Ethernet (registered trademark) differs from SONET / SDH, which has been developed as a network technology, in terms of granularity.
The situation is that many users do not use the full bandwidth of 10Gb / s even though the interface of 10 Gigabit Ethernet (registered trademark) has become widespread and each data channel has a bandwidth capacity of 10Gb / s physically Is also expected. From the user's standpoint, it is rather desirable to provide granularity between Gigabit Ethernet® and 10 Gigabit Ethernet®.
For example, SONET / SDH synchronization frame STM-n (n = 1, 4, 16, 64), such as hierarchical granularity, or virtual concatenation virtual by SONET / SDH virtual concatenation (virtual concatenation) It is desirable to provide stepped granularity such as a container VC-4-Nv (N = 1,..., 256).

As a conventional method for transferring packets on SONET / SDH, there is POS (PPP over SONET / SDH, IETF RFC 2615). The use of POS has the advantage that the data transfer efficiency is good because the packet loss and the overhead due to the cell header are less than the method of dividing the packet into ATM (Asynchronous Transfer Mode) cells.
Nevertheless, since POS uses HDLC (High-level Data Link Control procedure) like framing, there is a problem that the effective transmission rate is halved in the worst case by so-called byte stuffing. In addition, in what is called Ethernet (registered trademark) over SONET, GFP (Generic Framing Procedure, ITU-TG.7041), LAPS (Link Access Procedure-SDH, ITU-T X.86), etc., There are also relatively new standards.
These are also called next-generation SONET (NG-SONET) technologies as a method for efficiently mapping variable-length frames such as Ethernet (registered trademark) to SONET / SDH frames.

However, when a new intermediate layer protocol (GFP, LAPS, etc.) is introduced for transferring Ethernet (registered trademark) traffic through a SONET / SDH network, there is a problem that the layer structure and processing become complicated. Therefore, it is desirable to map an Ethernet (registered trademark) frame to a SONET / SDH payload without performing encapsulation or the like using an intermediate layer protocol.
Since the encoded Ethernet (registered trademark) data stream itself has a code synchronization mechanism, it can be mapped to the SONET / SDH payload as it is. However, there is a trade-off that the effective data rate is lowered (for example, it is reduced by 20% in the 8B / 10B code) due to the coding redundancy.
In this respect, 10 Gigabit Ethernet (registered trademark) adopts 64B / 66B codes, and is more efficient. In the WAN PHY of 10 Gigabit Ethernet (registered trademark) intended to be extended to the WAN, this 64B / 66B code is directly poured into the SONET STS-192c payload.

  Conventionally, there are technologies as described in Patent Documents 1 to 3 regarding a configuration for transferring Ethernet (registered trademark) -based traffic through a SONET / SDH system.

In paragraphs 12, 35 to 37 and Table 1 of Patent Document 1, a plurality of virtual containers having a bit rate lower than the bit rate of the data are generated for the data to be transmitted, and the above transmission is performed by virtual concatenation. A configuration is described in which data to be transmitted is transmitted by a virtual linked virtual container including the plurality of virtual containers.
As data to be transmitted, Ethernet (registered trademark) of 10 Mb / s, 100 Mb / s, and 1 Gb / s are listed, and these can be used for finer granularity without introducing a new intermediate layer protocol. Efficiently by using virtual containers.
In short, the technical point of view of this patent document is that the rate of Ethernet (registered trademark) is fixed, and the granularity and number of virtual containers that constitute a virtual linked virtual container are adjusted, so that SONET / The SDH virtual connected virtual container rate is adapted to Ethernet (registered trademark). In other words, with the known technique according to this patent document, it is not possible to set more flexible granularity in Ethernet (registered trademark) as a client.

FIG. 6 of Patent Document 2 describes a configuration for mapping a 10 Gb / s Ethernet (registered trademark) data stream to OC-192. The feature of this patent document is that the Ethernet (registered trademark) data stream is compressed for mapping to OC-192 with a line rate of 9.95328 Gb / s (the payload is 9.58464 Gb / s). Removes the inter frame gap (IFG) of the Ethernet® data stream.
Also, for the purpose of using a common PMD (Physical Media Dependent) interface, the line rate is 10.3125 Gb / s (when using 64B / 66B encoding adopted in 10 Gigabit Ethernet (registered trademark)). Physical rate). In other words, the subject of this patent document is not 10 Gigabit Ethernet (registered trademark).
In short, the technical point of view of this patent document is simply to map a data stream in the Ethernet® frame format to an interface having a line rate of 9.95328 Gb / s. Therefore, the granularity of Ethernet (registered trademark) cannot be made variable.

In Patent Document 3, FIGS. 8 and 10 to 13, a gigabit Ethernet (registered trademark) packet frame is added to the payload of a plurality of SONET synchronization frames in units of bytes, in units of the number of bytes of the payload, or in packet frames. A configuration for mapping in units of bytes is described. By mapping in units of bytes, padding is reduced and the payload is used effectively. Also, mapping is performed in units of the number of bytes of the payload or in units of the number of bytes of the packet frame, thereby reducing the number of mappings and enabling high-speed operation.
However, when mapping, it is necessary to encapsulate in HDLC-like frames for packet separation, so this is done with LAPS. In other words, an intermediate layer protocol must be introduced. Also, when mapping in units of the number of bytes of the packet frame, that is, mapping for each packet, in order to absorb the difference between the effective rate of the data stream of the packet to be mapped and the virtual container, A buffer must be provided in the previous stage.
JP 2000-115106 A (particularly paragraphs 12, 35 to 37 and Table 1) Japanese Patent Laid-Open No. 2001-274824 (particularly FIG. 6) Japanese Patent Laid-Open No. 2003-69519 (particularly FIG. 8, 10-13)

  In order to transfer Ethernet (registered trademark) traffic over the SONET / SDH network, we would like to provide connections with various granularities like SONET / SDH using a 10 Gigabit Ethernet (registered trademark) interface. In other words, an Ethernet (registered trademark) connection with flexible granularity of 10 Gb / s or less is formed on 10 Gigabit Ethernet (registered trademark), and these are connected to SONET / SDH line interfaces (OC-192 and OC-768). I want to map to the tributary that composes

  In addition, the method of mapping packet-based traffic such as Ethernet (registered trademark) to SONET / SDH shall be applied to the internal configuration of routers and other network devices, and the router backplane interface shall be standardized. Would like to contribute to reducing the cost of routers by realizing multi-source parts procurement in router development.

In SONET / SDH, byte interleaving / byte disinterleaving is performed for multiplexing / dividing tributaries. This is because in the SONET / SDH network operating in synchronization with the network, the mapping of the data stream is sequentially processed for each byte without using a buffer or the like. Even when data such as Ethernet (registered trademark) is mapped to multiple SONET / SDH virtual containers (including via virtual concatenated virtual containers), the data stream is mapped by being byte-interleaved into the virtual container group. The
However, a data stream originally formed by a series of byte sequences, such as an Ethernet frame, etc., when byte deinterleaved, the consecutive bytes are mapped separately into different virtual containers. End up. Therefore, the Ethernet® frame cannot be mapped as it is to a specific virtual container. That is, the Ethernet (registered trademark) data stream cannot correspond to a specific SONET / SDH tributary.
In order to map such a byte string (frame or packet) to the same virtual container, conventionally, as in Patent Document 3, means for providing a buffer for each corresponding virtual container has been required.

  The present invention has been made in view of the above problems, and the encoded digital data signal, such as 10 Gigabit Ethernet (registered trademark), is fractionated into subchannels having variable granularity, An object of the present invention is to provide a data processing method and a data processing apparatus that map the subchannel to a tributary of a SONET / SDH path by sequential processing, that is, by cut-through instead of store-and-forward.

  The present invention includes a first step of inputting at least one digital data signal, a second step of generating at least one digital data stream, and data included in the digital data signal input in the first step. A third step of arranging frames in order for each attribute of the data frame on the digital data stream generated in the second step according to the attribute of the data frame; and the data in the third step. A digital data sequence including the data frame having at least one of the same attributes as the digital data stream in which the frames are arranged becomes a byte sequence having a set length X (X is a positive integer) bytes. In the fourth step of adding idle data, and idle data in the fourth step A fifth step of mapping the digital data stream to which X is added to X virtual containers by byte disinterleaving without being scrambled, and the X virtual containers mapped in the fifth step And a sixth step of mapping each of these to at least one virtual container by byte interleaving (claim 1).

  By rearranging the order of the frames of the digital data signal according to the attributes of the frames and arranging the frames on the digital data stream, it is possible to logically form subchannels by time division on the digital data stream.

  Also, after adding idle data to this digital data stream appropriately so that each subchannel becomes X bytes, mapping to X virtual containers with byte disinterleave and then mapping to virtual containers with byte interleaving By doing this, X bytes once distributed and mapped to X virtual containers (belonging to the same subchannel) are mapped to the same virtual container, and this series of mapping is realized as sequential processing. Is done.

  The present invention further includes a seventh step of encoding the digital data stream to which idle data is added in the fourth step, wherein the number of code bits of the encoding and the byte are set as the integer X. 2. The data processing method according to claim 1, wherein an integer obtained by dividing an integer multiple of a least common multiple of the number of bits by the number of bits of the bytes is used (claim 2).

If the digital data stream is encoded, by selecting an integer X as described above, separated bytes of subchannels they match the code separated coded digital data stream, the SO NET / SDH It can be made compatible with the tributary multiplexing / division method (that is, byte interleaving / byte disinterleaving).
At this time, the digital data stream is unscrambled. This prevents bits from being mixed across byte delimiters and mapped to different virtual containers during byte disinterleaving.

  Further, the present invention provides an eighth mapping in which the at least one virtual container mapped in the fifth step is mapped to X virtual containers by byte disinterleaving in order for each of the at least one virtual container. And a ninth step of mapping and outputting each of the X virtual containers mapped from the at least one virtual container in the eighth step to a digital data stream by byte interleaving. 3. A data processing method according to claim 1 or 2, characterized by comprising (Claim 3).

  As a result, the digital data streams mapped to at least one virtual container can be collectively restored to one digital data stream.

  The present invention further includes an eighth step of mapping and outputting each of the at least one virtual container mapped in the fifth step to a digital data stream. Or a data processing method according to claim 2 (claim 4).

  Thereby, the digital data stream mapped to at least one virtual container can be individually restored as a digital data stream. Also, these digital data streams may be mixed with different granularities.

  In the tenth aspect of the present invention, the digital data stream in which the data frames are arranged is selected according to the attribute of the data frame among the at least one digital data stream generated in the second step. 5. The data processing method according to claim 1, further comprising a step (claim 5).

  Thereby, since a subchannel can be formed on the interface of a digital data stream having a plurality of different routes, a plurality of subchannels having the same input interface can be distributed to these routes.

  Further, the present invention provides an eleventh step of arranging the data frames in order for each of the attributes on the digital data stream generated in the second step and inserting a synchronization signal; The synchronization signal is extracted from the digital data stream in which the synchronization signal is inserted in the step, and at least one digital data sequence including the data frame having the same attribute is an X-byte byte sequence. 6. The data processing method according to claim 1, further comprising a twelfth step of adding idle data to (Claim 6).

As a result, when performing the steps of forming a subchannel (arranging frames in order according to attributes) and the step of aligning the length of the subchannel byte sequence to X bytes, the subchannel byte sequence is performed. In the step of aligning the length of the sub-channel to X bytes, it is possible to recognize the subchannel break by detecting the synchronization signal (without examining all the contents of the digital data stream).
As the “synchronization signal”, any bit pattern that cannot be generated as a serial bit string in the digital data stream may be used. Here, “extracting the synchronization signal” means “detecting the synchronization signal, establishing synchronization, and then extracting the synchronization signal”.

  Further, the present invention provides idle data such that the digital data stream generated in the second step includes a digital data string including at least one data frame having the same attribute as an X-byte byte string. And adding the synchronization signal, extracting the synchronization signal from the digital data stream in which the synchronization signal was inserted in the eleventh step, and mapping it to the X virtual containers The data processing method according to any one of claims 1 to 5, further comprising a twelfth step (claim 7).

Thereby, when mapping from the digital data stream to X virtual containers, it is possible to easily detect the first byte that starts to be mapped to the virtual container group.
Here, any bit pattern that cannot be generated as a serial bit string in the digital data stream may be used as the “synchronization signal”. “Extracting a synchronization signal” refers to “detecting a synchronization signal to establish synchronization, and then extracting the synchronization signal”.

  Further, the present invention is the data processing method according to claim 6 or 7, wherein a synchronization signal is inserted for each of the attributes of the data frame in the eleventh step (claim 8).

  As a result, synchronization can be established for each sub-channel in the digital data stream.

  Further, the present invention is the data processing method according to claim 6 or 7, wherein a synchronization signal is inserted in each cycle of the attribute of the data frame in the eleventh step (claim 9). ).

  As a result, synchronization can be established for each subchannel group in the digital data stream.

  Further, the present invention provides a thirteenth step of inserting n (n is a positive integer) bytes of stuff operation control bytes for every X bytes of the digital data stream generated in the second step, A fourteenth step in which n virtual containers are generated in addition to the X virtual containers to which the digital data stream is mapped in step 5 to form (X + n) virtual containers; The fifteenth step of mapping the digital data stream in which the control bytes for stuff operation are inserted in the step of (x + n) virtual containers generated in the fourteenth step by byte disinterleaving, and the fourteenth step Each of the X virtual containers mapped in the step is at least one virtual container. A sixteenth step of mapping by byte interleaving, a seventeenth step of mapping each of the n virtual containers generated in the fourteenth step to at least one virtual container by byte interleaving, An eighteenth step of performing a negative stuff operation on the at least one virtual container mapped from the n virtual containers using the staff operation control byte inserted in the thirteen steps; and the eighteenth step A negative stuffing operation for generating an information bit and a 20th step of transmitting a link layer control signal using the information bit generated in the nineteenth step. According to any one of claims 1 to 9, characterized in A chromatography data processing method (claim 10).

  Thereby, a control signal related to the link of the digital data stream can be transmitted by SONET / SDH separately from the subchannel group as a client signal of SONET / SDH.

  In the present invention, when mapping the X virtual containers mapped with the digital data stream in the fifth step to the at least one virtual container in the sixth step by byte interleaving, A thirteenth step of inserting a staff operation control byte into each of at least one virtual container, and at least one of at least one virtual container in which the staff operation control byte is inserted in the thirteenth step A 14th step of performing a negative stuff operation on a single virtual container, a 15th step of generating an information bit by the negative stuff operation of the 14th step, and an information bit generated in the 15th step Is used to transmit the link layer control signal. The data processing method according to any one of claims 1 to 9, further comprising: a sixteenth step to reach (claim 11).

  As a result, a control signal related to the data link can be defined and transmitted to each of at least one virtual container using SONET / SDH.

  12. The method according to claim 1, further comprising a twenty-first step in which an input interface number of the digital data signal to which the data frame is input is the attribute of the data frame. A data processing method according to any one of the items (claim 12).

  Thus, when there are a plurality of digital data signal input interfaces, each of the input interfaces can correspond to a subchannel.

  12. The method according to claim 1, further comprising a twenty-first step in which an output interface number of the digital data stream from which the data frame is output is used as the attribute of the data frame. A data processing method according to any one of the items (claim 13).

  Thereby, when there are a plurality of output interfaces for the digital data stream, each output interface can correspond to a subchannel.

  The present invention further includes a twenty-first step in which a VLAN tag of an Ethernet (registered trademark) frame in the input data frame or a part of the VLAN tag is used as the attribute of the data frame. A data processing method according to any one of claims 1 to 11 (claim 14).

  Thereby, when the digital data signal is Ethernet (registered trademark), it is possible to associate the subchannel with the tag VLAN of Ethernet (registered trademark). Here, the “VLAN tag” may be an extended VLAN tag such as “VMAN method” (Extreme Networks) or “802.10-in-Q method” (Cisco Systems).

  The present invention further includes a twenty-first step in which the MAC address of the Ethernet (registered trademark) frame in the input data frame or a part of the MAC address is used as the attribute of the data frame. A data processing method according to any one of claims 1 to 11 (claim 15).

  As a result, when the digital data signal is Ethernet (registered trademark), the subchannel can be associated with the Ethernet (registered trademark) link.

  The present invention further includes a twenty-first step in which an IP address of an IP packet in the input data frame or a part of the IP address is used as the attribute of the data frame. A data processing method according to any one of items 1 to 11 (claim 16).

  Thereby, when the digital data signal includes an IP packet, the subchannel can be associated with the IP address.

  The present invention further comprises a twenty-first step in which a Shim header label or a part of the Shim header label in the input data frame is used as the attribute of the data frame. The data processing method according to any one of items 1 to 11 (claim 17).

  As a result, when the digital data signal includes a packet with a MPLS (Multi Protocol Label Switching) label, the subchannel can be associated with the MPLS LSP (Label Switched Path).

  The present invention also provides an input interface number of the digital data signal to which the data frame is input, an output interface number of the digital data stream to which the data frame is output, and an Ethernet ( (Registered trademark) VLAN tag of the frame or part of the VLAN tag, MAC address of the Ethernet (registered trademark) frame in the input data frame or part of the MAC address, and in the input data frame Combining at least two of the IP address of the IP packet or a part of the IP address and the label of the Shim header or the part of the Shim header in the input data frame, the data frame 12. The method according to claim 1, further comprising a 21st step as an attribute. A data processing method according (claim 18).

  Thereby, the corresponding unit of the subchannel can be set more finely.

  According to the present invention, there is provided a data processing device including a first format conversion device and a second format conversion device, wherein the first format conversion device is a digital data signal input means for inputting at least one digital data signal. First digital data stream generation means for generating at least one digital data stream; and data frames in the digital data signal input from the digital data signal input means in order according to the attributes of the data frames, Data frame arranging means arranged on the digital data stream generated by the first digital data stream generating means, the second format conversion device, the at least one data frame having the same attribute is Including digital data strings set First idle data adding means for adding idle data so as to be a byte sequence of length X (X is a positive integer) bytes, and a digital data stream in which the data frame arrangement means arranges data frames At least one of each of the first mapping means for mapping X virtual containers by byte disinterleaving without being scrambled, and each of the X virtual containers mapped by the first mapping means A data processing apparatus comprising second mapping means for mapping to a virtual container by byte interleaving (claim 19).

  Thereby, formation of the subchannel (first format conversion device) and mapping of the subchannel to the virtual container (second format conversion device) can be realized.

  The present invention further includes digital data stream encoding means for encoding a digital data stream in which the data frame arrangement means has arranged the data frames, and the digital data stream encoding means is used as the integer X. 20. The data processing apparatus according to claim 19, wherein an integer obtained by dividing an integer multiple of a least common multiple of the number of code bits encoded by and the number of bits of the bytes by the number of bits of the bytes is used ( Claim 20).

If the digital data stream is encoded, by selecting an integer X as described above, separated bytes of subchannels matches the code separated coded digital data stream, the SO NET / SDH It can be made compatible with the tributary multiplexing / division method (that is, byte interleaving / byte disinterleaving).

  Further, the present invention provides a third mapping in which at least one virtual container mapped by the second mapping means is sequentially mapped to X virtual containers by byte disinterleaving for each of the at least one virtual container. A third format conversion apparatus comprising: a mapping means; and a fourth mapping means for mapping each of the X virtual containers mapped by the third mapping means to a digital data stream by byte interleaving. 21. The data processing device according to claim 19 or 20, wherein the data processing device is provided (claim 21).

  As a result, the digital data streams mapped to at least one virtual container can be collectively restored to one digital data stream.

  The present invention further comprises a third format conversion device having a third mapping means for mapping at least one virtual container mapped by the second mapping means to a digital data stream. A data processing device according to claim 19 or 20 (claim 22).

  Thereby, the digital data stream mapped to at least one virtual container can be individually restored as a digital data stream.

  23. The data according to claim 21, further comprising at least one SONET / SDH switch chip that connects the second format conversion device and the third format conversion device. A processing device (claim 23).

  As a result, a SONET / SDH switch chip is used as a switch fabric to configure a router, and the router backplane interface can be a standardized SONET / SDH interface.

  The data processing according to claim 21 or 22, further comprising at least one SONET / SDH device for connecting the second format conversion device and the third format conversion device. A device (claim 24).

  Thereby, the virtual container to which the subchannel is mapped can be transferred through the SONET / SDH network configured by the SONET / SDH device.

  The present invention is the data processing device according to claim 24, wherein the at least one SONET / SDH device further includes at least one SONET / SDH cross-connect device (claim 25).

  As a result, the virtual container to which the subchannel is mapped can be cross-connected (time slot exchange) by the SONET / SDH cross-connect in the SONET / SDH network.

  Further, the present invention provides the second format conversion apparatus, wherein the digital data stream includes idle data such that a digital data sequence including the data frame having the same attribute in at least one of the digital data streams is an X-byte byte sequence. The second idle data adding means for adding the idle data and the digital data stream to which the second idle data adding means has added the idle data without being scrambled to the X virtual containers by byte disinterleaving. 26. The data processing device according to claim 19, further comprising fifth mapping means for mapping (claim 26).

  As a result, the length of the sub-channel byte string can be aligned to X bytes by the second format conversion device. Further, the digital data stream can be transmitted in a scrambled state between the first format conversion device and the second format conversion device.

  Further, the present invention provides the first format conversion device, wherein the digital data stream includes idle data such that a digital data sequence including the data frame having the same attribute in at least one of the digital data streams is an X-byte byte sequence. The second format conversion device further includes a second idle data adding means for adding the idle data to the digital data stream to which the second idle data adding means has added the idle data without being scrambled. 26. The data processing device according to claim 19, further comprising fifth mapping means for mapping to the X virtual containers by byte disinterleaving (claim 27). .

  As a result, the length of the sub-channel byte string can be aligned to X bytes by the first format conversion device. Further, the digital data stream can be transmitted in a scrambled state between the first format conversion device and the second format conversion device.

  In the present invention, the first format conversion device further includes second digital data stream generation means for generating the digital data stream without being scrambled, and the second format conversion device The idle data is added to the digital data stream generated by the second digital data stream generating means so that the digital data string including the data frame having the same at least one attribute is an X-byte byte string. The data processing device according to any one of claims 19 to 25, further comprising second idle data adding means for performing the processing (claim 28).

  As a result, the length of the sub-channel byte string can be aligned to X bytes by the second format conversion device. In addition, the second format conversion device can save time and effort to unscramble the digital data stream.

Further, the present invention provides the second digital data stream generation means, wherein the first format conversion device generates the digital data stream without being scrambled,
Idle data is added to the digital data stream generated by the second digital data stream generation means so that a digital data sequence including the data frame having the same attribute is at least one byte sequence. 26. The data processing device according to claim 19, further comprising second idle data adding means for performing the processing (claim 29).

  As a result, the length of the sub-channel byte string can be aligned to X bytes by the first format conversion device. In addition, the second format conversion device can save time and effort to unscramble the digital data stream.

  In the present invention, the first format conversion device further includes synchronization signal insertion means for inserting a synchronization signal into the digital data stream, and the second format conversion device extracts the synchronization signal. 30. The data processing device according to claim 19, further comprising a signal extraction unit (claim 30).

  As a result, the second format conversion device can recognize the subchannel delimiter by detecting the synchronization signal, and can easily detect the head byte. As the “synchronization signal”, any bit pattern that cannot be generated as a serial bit string in the digital data stream may be used. Here, “extracting the synchronization signal” means “detecting the synchronization signal, establishing synchronization, and then extracting the synchronization signal”.

  32. The data processing device according to claim 30, wherein the synchronization signal inserting means inserts the synchronization signal into the digital data stream for each of the attributes of the data frame. (Claim 31).

  Thus, the second format conversion device can synchronize each subchannel in the digital data stream.

  32. The data processing device according to claim 30, wherein the synchronization signal inserting means inserts the synchronization signal into the digital data stream for each round of the attribute of the data frame. (Claim 32).

  Thus, the second format conversion device can synchronize each subchannel group in the digital data stream.

  In the present invention, the first format converter further includes a FIFO queue for storing the data frame in the digital data signal input from the digital data signal input means for each attribute of the data frame. A data processing device according to any one of claims 19 to 32, characterized by comprising (claim 33).

  As a result, when subchannels are formed by sequential processing, it is possible to prevent the processing from being in time and discarding frames.

  The present invention is further characterized in that the first format conversion device further comprises a FIFO queue setting means for setting the depth and input / output speed of the FIFO queue for each attribute. A data processing apparatus according to claim 34 (claim 34).

  Thereby, the conditions regarding the band of a subchannel can be set, and each subchannel can be controlled separately.

  Further, the present invention provides the non-priority data frame detection means, wherein the first format conversion device detects a non-priority data frame among the data frames in the digital data signal input from the digital data signal input means. 35. The data processing apparatus according to claim 33, further comprising non-priority data frame discarding means for discarding the non-priority data frame without storing it in the FIFO queue. 35).

  Thus, frame priority control within one subchannel can be performed.

  Further, the present invention provides the non-priority data frame detection means, wherein the first format conversion device detects a non-priority data frame among the data frames in the digital data signal input from the digital data signal input means. 35. The data processing apparatus according to claim 33, further comprising non-priority data frame discarding means for outputting the non-priority data frame from the FIFO queue and immediately discarding the non-priority data frame. 36).

  Thereby, similarly, priority control of frames within one subchannel can be performed.

  In the present invention, the non-priority data frame detection means may include non-priority data frame identification means for identifying the non-priority data frame from information mounted on an IP header or a Shim header of an IP packet in the data frame. 37. The data processing device according to claim 35 or 36, further comprising (claim 37).

  Thereby, frame priority control within the subchannel can be performed with reference to the priority in the IP layer or MPLS layer.

  According to the data processing method and the data processing apparatus of the present invention configured as described above, virtual sub-sequences are sequentially performed every X bytes on a serial digital data stream such as 10 Gigabit Ethernet (registered trademark). A channel is formed and mapped to X virtual containers (with scrambled) with byte disinterleaving, and then mapped to at least one virtual container with byte interleaving, so that X virtual Since each byte of the subchannel distributed in the container can be concentrated and mapped to at least one virtual container, the remarkable effect of cutting through the subchannel to the SONET / SDH tributary (mapping in sequential processing) There is an effect.

  Here, in the subchannel, about 50 Mb / s and its integral multiple (VC-3-Nv), about 150 Mb / s and its integral multiple (VC-4-Nv), etc. according to the bandwidth of the virtual container. , It can have flexible granularity.

  When the SONET / SDH switch chip is used, the part corresponding to the interface between the line cards of the network device that exchanges packets such as routers can be connected via the SONET / SDH tributary. . This apparatus realizes a function equivalent to a packet switch by setting a connection destination of a SONET / SDH tributary for each destination of a packet included in a digital data stream. That is, the backplane interface in the router can be a standardized interface that does not include any unique format.

Further, when the SONET / SDH transmission apparatus is provided, a virtual container in which subchannels are mapped can be transferred on a SONET / SDH network configured by the SONET / SDH transmission apparatus. That is, packet-based traffic such as Ethernet (registered trademark) can be transferred over the SONET / SDH network with flexible granularity.
When the SONET / SDH cross-connect is used, the virtual container to which the subchannel is mapped can be cross-connected (time slot exchange) with the SONET / SDH cross-connect. That is, it is possible to control the route on the SONET / SDH network while transferring the packet-based traffic having flexible granularity on the SONET / SDH network.

  The best mode for carrying out the present invention will be described below with reference to FIGS.

  FIG. 1 shows the concept of a data processing method (mapping) according to an embodiment of the present invention.

A data frame is input by the digital data signal 110. In this example, these data frames are input in the order of 311, 321, 341, 331, 332 and 322. The gap between these data frames is filled with an interframe gap (not shown) when the digital data signal 110 is, for example, 10 Gigabit Ethernet (registered trademark).
Here, the data frame 311, the data frame 321, the data frame 331, and the data frame 341 are data frames belonging to different attributes. The data frame 322 is a data frame belonging to the same attribute as the data frame 321, and the data frame 332 is a data frame belonging to the same attribute as the data frame 331.

These data frames are arranged in order for each attribute on the digital data data stream 120 to form subchannels 211, 212, 213, and 214. That is, subchannels are formed on the digital data stream 120 in a time division manner. Therefore, if the number of subchannels is n, each subchannel repeatedly appears as a byte string every n on the digital data stream.
Data frame 311 is included in subchannel 211, data frames 321 and 322 are included in subchannel 212, data frames 331 and 332 are included in subchannel 213, and data frame 341 is included in subchannel 214, respectively. In addition, here, idle data (not shown) is added so that the digital data string of each subchannel is a byte string of a certain set X (X is a positive integer) bytes.

  Each subchannel is then mapped to a virtual container. The subchannel 211 is mapped to the virtual container 411, the subchannel 212 is mapped to the virtual container 412, the subchannel 213 is mapped to the virtual container 413, and the subchannel 214 is mapped to the virtual container 414. At this time, the digital data stream 120 is mapped to each virtual container without being scrambled.

  FIG. 2 is a diagram showing a concept of a data processing method (demapping) according to the embodiment of the present invention.

Each virtual container to which the subchannel is mapped is demapped to the digital data stream. On the digital data stream 130, the subchannels 211, 212, 213, and 214 corresponding to the virtual containers 411, 412, 413, and 414, respectively, are restored as byte strings (having a length of X bytes).
Subchannel 211 includes data frame 311, subchannel 212 includes data frames 321 and 322, subchannel 213 includes data frames 331 and 332, subchannel 214 includes data frame 341, and idle data is included as appropriate. Will be restored.

  FIG. 3 is a diagram showing a concept of a data processing method (demapping for each VC) according to the embodiment of the present invention.

Each virtual container 411, 412, 413, 414 to which the subchannel is mapped is demapped to a digital data stream 131, 132, 133, 134, respectively. On the digital data stream 131, the subchannel 211 corresponding to the virtual container 411 is restored as a byte string (having a length of X bytes) so that the data frame 311 is included.
On the digital data stream 132, the subchannel 212 corresponding to the virtual container 412 is restored as a byte sequence (length X bytes) in such a way that the data frames 321 and 322 are included.
On the digital data stream 133, the subchannel 213 corresponding to the virtual container 413 is restored as a byte sequence (having a length of X bytes) so that the data frames 331 and 332 are included.
On the digital data stream 134, the subchannel 214 corresponding to the virtual container 414 is restored as a byte sequence (length X bytes) in a form including the data frame 341.
In this case, each sub-channel is restored as a digital data stream including idle data as appropriate.

  FIG. 4 shows in detail the mapping by byte disinterleaving / byte interleaving from a digital data stream in which subchannels are formed according to an embodiment of the present invention to a (virtually concatenated) virtual container.

On the digital data stream 120, sub-channel byte sequences 211, 212, 213, and 214 are arranged. Each of these byte strings has a length of X bytes. The digital data stream 120 in which the subchannels are formed is mapped to the container 420 as a byte sequence, and is mapped to X virtual containers 430-1, 430-2,..., 430-X by byte disinterleaving.
At this time, the digital data stream 120 is mapped in a scrambled state. As a result, the byte strings of the subchannels 211, 212, 213, and 214 are distributed byte by byte in the X virtual containers 430-1 to X.

Subsequently, the virtual containers 430-1 to 430-X are further mapped to the containers 441, 442, ... (hereinafter not shown) by byte interleaving. As a result, the byte sequences of the subchannels 211, 212,... Distributed in the X virtual containers 430-1 to 430-X are mapped to the containers 441, 442,.
Each of the containers 441, 442,... Corresponds to a (virtually connected) virtual container 411, 412,... And is mapped to a virtual container group constituting the (virtually connected) virtual container. Therefore, the subchannels 211, 212,... Are mapped to (virtually connected) virtual containers 411, 412,.
The handling of the lOGbE subchannels 213 and 214 is not shown here, but is the same as that of the subchannels 211 and 212. Also, here, the virtual containers 411, 412,... May be virtual linked virtual containers or virtual containers.

  Further, when mapping from the digital data stream 120 to the virtual containers 430-1 to 430 -X, the procedure via the container 420 may be realized sequentially. Therefore, the container 420 may not actually prepare a buffer having a capacity corresponding to C-4-Xc. The same applies to the containers 441, 442,.

  FIG. 5 shows the details of demapping by byte interleaving / byte disinterleaving from a virtual container with mapped subchannels (virtual concatenation) to a digital data stream according to an embodiment of the present invention.

(Virtually connected) Sub-channels 211, 212,... Are mapped to virtual containers 411, 412,. The (virtually connected) virtual containers 411, 412, ... are mapped to the containers 451, 452, ... from the virtual container group constituting each (virtually connected) virtual container.
Further, the containers 451, 452,... Are mapped to X virtual containers 460-1, 460-2,. As a result, the byte strings of the subchannels 211, 212,... Are distributed to the X virtual containers 460-1 to X. The handling of the subchannels 213 and 214 is not shown here, but is the same as that of the subchannels 211 and 212.

Subsequently, the virtual containers 460-1 to 460 -X are further mapped to the container 470 by byte interleaving. As a result, the byte strings of the subchannels 211, 212, 213, and 214 distributed to the X virtual containers 460-1 to 460-X are concentrated on the container 470 and mapped as an X-byte byte string.
Container 470 is mapped to digital data stream 130. On the digital data stream 130, X-byte byte sequences 211, 212, 213, and 214 are restored as byte sequences.

  Note that when demapping from the virtual containers 460-1 to X to the digital data stream 130, the procedure via the container 470 may be realized sequentially. Therefore, the container 470 may not actually prepare a buffer having a capacity corresponding to C-4-Xc. The same applies to the containers 451, 452,.

As described above, the subchannel is formed as a byte sequence of X bytes on the digital data stream. Subsequently, the digital data stream in which the sub-channel is formed is mapped to X virtual containers by byte disinterleaving (in a scrambled state), and further mapped to the virtual container by byte interleaving (virtual concatenation).
This maps the subchannel to the (virtual concatenation) virtual container. Further, these (virtually linked) virtual containers are mapped to X virtual containers by byte interleaving, and further mapped to the digital data stream by byte disinterleaving. This demaps the subchannel from the (virtual concatenation) virtual container. As described above, the subchannel can be associated with the SONET / SDH tributary by cut-through (sequential processing).

  FIG. 6 illustrates details of demapping by byte interleaving / byte disinterleaving into a digital data stream for each virtual container to which subchannels are mapped (virtual concatenation) according to an embodiment of the present invention.

(Virtually connected) Sub-channels 211, 212,... Are mapped to virtual containers 411, 412,. The (virtually connected) virtual containers 411, 412, ... are mapped to the containers 451, 452, ... from the virtual container group constituting each (virtually connected) virtual container.
Demapped from containers 451, 452,... To digital data streams 131, 132, respectively. The byte sequence of the subchannel 211 is restored on the digital data stream 131, and the byte sequence of the subchannel 212 is restored on the digital data stream 132, respectively.

As described above, on the side of demapping to the subchannel digital data stream, the virtual container corresponding to each subchannel is individually mapped to the digital data stream to restore the data frame of the original subchannel. Can do.
The restored digital data stream is only added with idle data as appropriate, and is completely compatible with the original digital data stream. Therefore, the demapping side does not need to perform byte disinterleave / byte interleave processing, and this enables application development to unidirectional data transmission, multicast transfer that is expanded from the data transmission, and the like.

  FIG. 7 shows an example of 10 gigabit Ethernet (registered trademark) regarding the principle of matching between encoding and byte multiplexing according to an embodiment of the present invention, that is, the length adjustment of a byte sequence of a subchannel.

  For 10 Gigabit Ethernet (registered trademark), 64B / 66B codes are used. Therefore, the integer X is obtained by dividing the integer multiple of the least common multiple 264 of the number of bits of the code 66 and the number of bits of bytes (= octets) 8 in the communication field by the number of bits of bytes 8 and is a multiple of 33. It will be.

FIG. 7 shows an enlarged view of the data frame 311 (including the front and back) of the digital data signal 110. Here, it is assumed that the digital data signal 110 is a 10 Gigabit Ethernet (registered trademark) signal, and the data frame 311 is an Ethernet (registered trademark) frame.
The digital data signal 110 (of 10 Gigabit Ethernet (registered trademark)) is shown in a state where the 64B payload of the 64B / 66B code is unscrambled. The digital data signal 110 (of 10 Gigabit Ethernet (registered trademark)) includes p 64B / 66B codes from 64B / 66B codes 630-1, 630-2 to 630-p. Each 64B / 66B code includes a 2-bit header 631-1, 631-2, ..., 631-p and a 64B payload 632-1, 632-2, ..., 632-p.
The data frame 311 (of Ethernet®) is included in these p 64B / 66B codes, and the 64B payload 632-1 of the 64B / 66B code 630-1 is (of Ethernet®) The head part (indicated by / s /) of the data frame 311 is included, and the 64B payload 632-p of the 64B / 66B code 630-p is the end part (/ t / of the Ethernet®). And the other 64B payload between them contains data (indicated by / d /).
Here, 64B / 66B code 630-1 and 630-p are control blocks in which control columns and data columns are mixed, so 64B payloads 632-1 and 632-p are actually 8-bit block types. It consists of a field (indicating a combination of control column and data column) and a 56-bit data field. The 64B / 66B code 630-2 is a data block consisting of all data, and the 64B payload 632-2 is composed of 8 8-bit data bytes.

  The digital data signal 110 (of 10 Gigabit Ethernet (registered trademark)) is represented as a column column 601 in XGMII (10 Gigabit Media Independent Interface) by removing the 2-bit header and dividing it into four lanes. The column row 601 includes two column blocks 610-1, 610-2, ..., 610-p. The two column blocks 610-1, 610-2, ..., 610-p correspond to 64B / 66B codes 630-1, 630-2, ..., 630-p, respectively.

  An idle column (all / i / columns) is added to the column row 601 to form a column row 602. At this time, if the length of the column row 602 is (X × 8) ÷ 66 × 2 rows, when this is expanded into a 64B / 66B code, the whole becomes exactly X bytes. In this example, the idle column at the beginning of the column row is removed, and the idle column is added to the end of the column row. As a result, the column row 602 is a column row composed of q two column blocks of two column blocks 620-1, 620-2,..., 620-q. At this time, the relationship of p ≦ q is established between the integers p and q.

This column column consisting of q 2 column blocks 620-1, 620-2, ..., 620-q is a digital data stream 120 (for Ethernet (registered trademark)) 120 (Ethernet (registered trademark) frame for each attribute) In the above, 64B / 66B codes 640-1, 640-2,..., 640-q are obtained. Here, the 64B / 66B code 64B payload is unscrambled. 64B / 66B codes 640-1, 640-2, ..., 640-q are 64B payloads 642-1, 642-2 with 2-bit headers 641-1, 641-2, ..., 64-1-q added, respectively. ..., 642-q.
64B / 66B codes 640-1, 640-2,..., 640-q include a data frame 311 (of Ethernet®), and the total length is X bytes. Here, the integer X is a multiple of 33. Thereby, the X byte corresponding to one subchannel, that is, the subchannel delimiter can be matched with the byte delimiter and the code delimiter.
Here, 64B / 66B codes 640-1 and 640-q are control blocks in which control columns and data columns are mixed, so the 64B payloads 642-1 and 642-q are actually 8-bit block types. It consists of a field (indicating a combination of control column and data column) and a 56-bit data field. The 64B / 66B code 640-2 is a data block consisting of all data, and the 64B payload 642-2 is composed of eight 8-bit data bytes.

  Here, the explanation is focused on digital data streams such as 10 Gigabit Ethernet (registered trademark) and one-way transmission in SONET / SDH, but 10 Gigabit Ethernet (registered trademark) and SONET / SDH are Of course, the transmission is bidirectional, and the operation in the opposite direction is the same, and the same applies to the following description.

FIG. 8 is a conceptual diagram showing a basic example of the configuration of the data processing device (mapping) according to the embodiment of the present invention. Data frames 311, 321, 341, 331, 332, 322 on the digital data signal 110 input to the data processor 501 are fixed length (X bytes) on the digital data stream 120 by the first format converter 510. ) Are arranged in order for each attribute, and subchannels 211, 212, 213, and 214 are formed.
The byte sequences of the subchannels 211, 212, 213, and 214 formed on the digital data stream 120 are scrambled and the second format conversion device 520 performs (virtual concatenation) virtual containers 411, 412, 413 and 414 are mapped. These (virtually connected) virtual containers constitute a SONET / SDH link 401 as a tributary.

FIG. 9 is a conceptual diagram showing a basic example of the configuration of the data processing device (demapping) according to the embodiment of the present invention. (Virtually connected) The virtual containers 411, 412, 413, 414 are demapped to the digital data stream 130 by the third format conversion device 530 of the data processing device 502.
On the digital data stream 130, the subchannels 211, 212, 213, and 214 are restored as byte sequences of X bytes, and the data frames 311, 321, 322, 331, 332, and 341 are included in the byte sequences. Restored in the form.

FIG. 10 is a conceptual diagram showing a basic example of the configuration of the data processing apparatus (demapping for each VC) according to the embodiment of the present invention. (Virtually connected) Virtual containers 411, 412, 413, and 414 are demapped to digital data streams 131, 132, 133, and 134 by third format converters 5301, 5302, 5303, and 5304 of data processing device 502, respectively. The
On the digital data stream 131, the subchannel 211 is restored as a sequence of X byte sequences, and the data frame 311 is restored in the form of being included in these byte sequences.
On the digital data stream 132, the sub-channel 212 is recovered as a sequence of X byte sequences, and the data frames 321 and 322 are recovered in the form of inclusion in those byte sequences.
On the digital data stream 133, the sub-channel 213 is restored as a sequence of X byte sequences, and the data frames 331 and 332 are restored in the form of those byte sequences.
On the digital data stream 134, the subchannel 214 is recovered as a sequence of X byte sequences, and the data frame 341 is recovered in the form of inclusion in these byte sequences.

The length X of each subchannel's byte sequence is the integer multiple of the least common multiple of the number of sign bits and the number of bytes of bytes divided by the number of bits of bytes if the digital data stream is encoded. By using this integer, it is possible to provide compatibility with encoding and byte multiplexing (of SONET / SDH).
Further, the length X of the byte string needs to be large enough to include at least one data frame (variable length) having a practical length. For example, in the case of 10 Gigabit Ethernet (registered trademark) adopting 64B / 66B codes, 1518 or the like is preferable. The SONET / SDH link should have a bandwidth of 10 Gb / s or higher, such as OC-192 / STM-64 (about lOGb / s) or OC-768 / STM-256 (about 40 Gb / s). Is preferred. Further, when using a virtual connected virtual container, it is possible to perform inverse multiplexing using a plurality of SONET / SDH links with smaller bandwidths.

FIG. 11 shows an example of selecting a digital data stream in which a data frame is arranged according to the attribute of the data frame when configuring the subchannel according to the embodiment of the present invention. The data frames 311, 331, and 321 on the digital data signal 111 and the data frames 322, 341, and 332 on the digital data signal 112 input to the data processing device 501 are converted into a digital data stream by the first format conversion device 510. Located on 121 and 122.
Here, the digital data stream in which the data frame is arranged is selected according to the attribute of each data frame. In other words, the data frame 311 is arranged on the digital data stream 121 in a form included in the byte string of the subchannel 211.
On the other hand, the data frames 321 and 322 are arranged on the digital data stream 122 so as to be included in the byte string of the subchannel 212.
The data frames 331 and 332 are included in the digital data stream 121 in the form of being included in the byte sequence of the subchannel 213, and the data frame 341 is included in the digital data stream 122 in the form of being included in the byte sequence of the subchannel 214. Be placed.
Further, the subchannels 211 and 213 are mapped to (virtually connected) virtual containers 411 and 413 by the second format conversion device 521, respectively. The subchannels 212 and 214 are mapped to (virtually connected) virtual containers 412 and 414 by the second format conversion device 522, respectively.
The (virtual connection) virtual containers 411 and 413 constitute a SONET / SDH link 402 as a tributary, and the (virtual connection) virtual containers 412 and 414 constitute a SONET / SDH link 403 as a truth.

  Here, the number of subchannels formed in different digital data streams may be different. In addition, the lengths of sub-channel byte sequences formed in different digital data streams may be different.

FIG. 12 shows a device configuration example in the case of including a SONET / SDH switch chip according to an embodiment of the present invention.
The procedure until data frames 311, 321, 322, 331, 332 341 are mapped to (virtually connected) virtual containers 411, 412, 413, 414 as subchannels 211, 212, 213, 214 is shown in FIG. Same as the case.
(Virtually connected) Virtual containers 411 and 413, 412 and 414 are connected to the destination port of each data frame via SONET / SDH switch chip 550 as tributaries of SONET / SDH links 402 and 403, respectively (virtually connected) Connected to the container. Here, (virtual connection) virtual container 411 is (virtual connection) virtual container 416, (virtual connection) virtual container 413 is (virtual connection) virtual container 418, and (virtual connection) virtual container 412 is (virtual connection) virtual container 416 A virtual container 414 (virtually linked) to the container 415 is connected to a virtual container 417 (virtually linked).
(Virtually connected) The virtual containers 415 and 416, 417 and 418 constitute SONET / SDH links 404 and 405 as tributaries, respectively. (Virtually connected) The virtual containers 415 and 416 are demapped onto the digital data stream 131 by the third format conversion device 531 of the data processing device 502 and restored as subchannels 211 and 212. In addition, the (virtually connected) virtual containers 417 and 418 are demapped onto the digital data stream 132 by the third format conversion device 532 of the data processing device 502 and restored as subchannels 213 and 214.

The configuration of the data processing devices 501 and 502 is expressed as a functional block. When the device is actually configured, these functions may be arranged in any manner on the device. Data processing devices 501, 502 and SONET / SDH switch chip 550 can be linked to realize a function equivalent to a packet switching processing device such as a router.
Devices including data processing devices 501, 502 and SONET / SDH switch chip 550 can also be configured in a chassis-based form factor, such as a typical router. In that case, the backplane interface is a standardized SONET / SDH interface. Therefore, the data processing apparatus having the configuration according to the embodiment of the present invention can easily follow the increase in the physical speed of the backplane.

FIG. 13 shows a device configuration example in the case of including a SONET / SDH device according to an embodiment of the present invention. The procedure for mapping data frames 311, 321, 322, 331, 332, 341 as subchannels 211, 212, 213, 214 to (virtually connected) virtual containers 411, 412, 413, 414 is shown in FIG. Same as the case.
(Virtually connected) Virtual containers 411 and 413, 412 and 414 are transferred via SONET / SDH devices 551 and 552 as tributaries of SONET / SDH links 402 and 403, respectively. Thereafter, the (virtually connected) virtual containers 411 and 413 are demapped onto the digital data stream 131 by the third format conversion device 531 of the data processing device 502 and restored as subchannels 211 and 213.
Also, the (virtually connected) virtual containers 412 and 414 are demapped onto the digital data stream 132 by the third format conversion device 532 of the data processing device 502 and restored as subchannels 212 and 214.

  Any number of SONET / SDH devices may be interposed between the above-mentioned SONET / SDH devices, and the subchannels are transferred on a SONET / SDH network including these SONET / SDH devices.

FIG. 14 shows an example of a device configuration when a SONET / SDH cross-connect according to an embodiment of the present invention is included.
The procedure for mapping data frames 311, 321, 322, 331, 332, 341 as subchannels 211, 212, 213, 214 to (virtually connected) virtual containers 411, 412, 413, 414 is as shown in FIG. The same. (Virtually connected) Virtual containers 411 and 413, 412 and 414 are transferred via SONET / SDH cross-connects 553 and 554 as tributaries of SONET / SDH links 402 and 403, respectively.
At the same time, it is cross-connected by SONET / SDH cross-connects 553 and 554, and the route is switched. That is, the (virtually connected) virtual containers 411 and 412 are tributaries of the SONET / SDH link 404, and the (virtually connected) virtual containers 413 and 414 are tributaries of the SONET / SDH link 405.
Thereafter, the (virtually connected) virtual containers 411 and 412 are demapped onto the digital data stream 131 by the third format conversion device 531 of the data processing device 502 and restored as subchannels 211 and 212. Further, the (virtually connected) virtual containers 413 and 414 are demapped onto the digital data stream 132 by the third format conversion device 532 of the data processing device 502 and restored as subchannels 213 and 214.

  Any number of SONET / SDH devices (including SONET / SDH cross-connects) may be interposed between the above SONET / SDH cross-connects, and the subchannel is a SONET / SDH device comprising these SONET / SDH devices. Transferred and cross-connected over the network. At that time, it is preferable to provide signaling means for the SONET / SDH cross-connect in the SONET / SDH network, and to perform cross-connection setting control by signaling using a protocol such as RSVP-TE GMPLS extension.

FIG. 15 shows an operation related to byte string length adjustment when the synchronization signal for each subchannel according to the embodiment of the present invention is used. The data frames 311, 321, 341, 331, 332, and 322 on the digital data signal 110 are sequentially arranged by attribute on the digital data stream 121 by the first format conversion device 510. At the same time, the synchronization signal 711, 712, 713, 714 is inserted for each sub-channel by the means 561 for inserting the synchronization signal.
The second format conversion device 520 recognizes the subchannel breaks by detecting the synchronization signals 711, 712, 713, 714 from the digital data stream 121 by means 562 for extracting the synchronization signal, and detects the synchronization signal 711, 712, 713, and 714 are deleted, and the subchannels 211, 212, 213, and 214 are formed on the digital data stream 122 as byte sequences of X bytes.

FIG. 16 shows operations related to byte string length adjustment and mapping head byte detection when using a synchronization signal for each subchannel and each subchannel group according to the embodiment of the present invention.
The data frames 311, 321, 341, 331, 332, and 322 on the digital data signal 110 are sequentially arranged by attribute on the digital data stream 121 by the first format conversion device 510. At the same time, synchronization signals 711, 712, 713, 714 and 721 are inserted for each sub-channel by means 561 for inserting synchronization signals. The second format conversion device 520 recognizes the subchannel breaks by detecting the synchronization signals 711, 712, 713, 714 from the digital data stream 121 by means 562 for extracting the synchronization signal, and detects the synchronization signal 711, 712, 713, and 714 are deleted, and the subchannels 211, 212, 213, and 214 are formed on the digital data stream 122 as byte sequences of X bytes.
The second format converter 520 also detects the synchronization signal 721 from the digital data stream 121 by means 562 for extracting the synchronization signal, thereby mapping the subchannel group to the (virtually connected) virtual container group. Recognizes the first byte.

FIG. 17 shows an operation related to the mapping head byte detection when the synchronization signal for each subchannel according to the embodiment of the present invention is used.
The data frames 311, 321, 341, 331, 332, and 322 on the digital data signal 110 are sequentially arranged for each attribute as a byte sequence of X bytes on the digital data stream 121 by the first format converter 510. Is done. At the same time, the synchronization signal 711, 712, 713, 714 is inserted for each sub-channel by the means 561 for inserting the synchronization signal.
The second format conversion device 520 detects the synchronization signals 711, 712, 713, 714 from the digital data stream 121 by means 562 for extracting the synchronization signals, and maps each subchannel to a (virtually connected) virtual container. Recognize the first byte.

FIG. 18 illustrates an operation related to mapping head byte detection when a synchronization signal for each subchannel group according to the embodiment of the present invention is used.
The data frames 311, 321, 341, 331, 332, and 322 on the digital data signal 110 are sequentially arranged for each attribute as a byte sequence of X bytes on the digital data stream 121 by the first format converter 510. Is done. At the same time, the synchronization signal 721 is inserted for each subchannel group by the means 561 for inserting the synchronization signal.
The second format conversion device 520 detects the synchronization signal 721 from the digital data stream 121 by the means 562 for extracting the synchronization signal, thereby mapping the subchannel group to the (virtually connected) virtual container group. Recognize bytes.

  As the synchronization signal, a bit pattern that does not occur as a serial bit string is used in the digital data stream. For example, if the digital data stream is 10 Gigabit Ethernet (registered trademark), a reserved pattern that is unused in the 10 Gigabit Ethernet (registered trademark) standard among the control codes in 7-bit representation of 64B / 66B code ("0101101", "0110011", etc.) can be used. In addition, a bit pattern having an arbitrarily defined length may be used. The above is the same for the synchronization signal in other cases.

  In addition, when both the synchronization signal for each subchannel and the synchronization signal for each subchannel group are used at the same time, it is necessary to use different bit patterns for both. At this time, with respect to the first subchannel, detection of subchannel breaks can be performed by a synchronization signal for each subchannel group, and therefore the synchronization signal for each subchannel group may be omitted.

FIG. 19 is a diagram illustrating an example of a device configuration according to an embodiment of the present invention, in which the length adjustment of the byte sequence of the subchannel is performed by the second format converter, and between the first format converter and the second format converter This shows the operation when the digital data stream in which the subchannel is formed is transferred in a scrambled state.
The data frames 311, 321, 341, 331, 332, 322 on the digital data signal 110 are converted into data frames 311, 321, 322 in order by attribute on the digital data stream 121 by the first format converter 510. , 331, 332, 341 are arranged in this order. The digital data stream 121 is transferred from the first format converter 510 to the second format converter 520 as a scrambled digital data stream 122. Here, means for scrambling and descrambling the digital data stream 122 are not shown.
The second format conversion device 520 adjusts the length of the byte string of each sub-channel by means 571 for adding idle data, and converts the sub-channels 211, 212, 213, and 214 into X-bytes on the digital data stream 123. Form as byte sequence. Also, the second format conversion device 520 uses the descrambling function 572 to map the digital data stream 123 to the (virtual concatenation) virtual container as unscrambled.

FIG. 20 is a diagram illustrating an example of a device configuration according to an embodiment of the present invention, in which the length adjustment of the byte sequence of the subchannel is performed by the first format converter, and between the first format converter and the second format converter The digital data stream in which the sub-channel is formed shows an operation in a case where the digital data stream is transferred in a scrambled state.
The data frames 311, 321, 341, 331, 332, 322 on the digital data signal 110 are converted into data frames 311, 321, 322 in order by attribute on the digital data stream 121 by the first format converter 510. , 331, 332, and 341, and by adjusting the length of the byte string of each subchannel by means 571 for adding idle data, subchannels 211, 212, 213, and 214 are placed on the digital data stream 121. It is formed as a byte sequence of X bytes.
The digital data stream 121 is transferred from the first format converter 510 to the second format converter 520 as a scrambled digital data stream 122. Here, means for scrambling and descrambling the digital data stream 122 are not shown.
The second format conversion device 520 uses the descrambling function 572 to map the digital data stream 123 to the (virtually connected) virtual container as the unscrambled state.

  In the above two examples, the digital data stream in the data processing device is scrambled and the DC balance of the signal is maintained, so it can be said that the configuration is suitable for transfer using an optical signal interface. As the optical signal interface, for example, VSR (Very Short Reach), which has been implemented as an IA (Implementation Agreement) by OIF (Optical Internetworking Forum), can be used. In addition, various optical transceivers (XENPAK, XPAK, X2) compliant with XAUI can be used.

FIG. 21 is a diagram illustrating an example of a device configuration according to an embodiment of the present invention, in which a subchannel byte string length adjustment is performed by a second format converter, and between the first format converter and the second format converter. The digital data stream in which the subchannel is formed shows an operation in the case where the digital data stream is transferred in a descrambled state.
The data frames 311, 321, 341, 331, 332, 322 on the digital data signal 110 are converted into data frames 311, 321, 322 in order by attribute on the digital data stream 121 by the first format converter 510. , 331, 332, 341 are arranged in this order. The digital data stream 121 is descrambled by the descrambling function 572 and transferred from the first format converter 510 to the second format converter 520.
The second format conversion device 520 adjusts the length of the byte string of each subchannel by means 571 for adding idle data, and places the subchannels 211, 212, 213, 214 on the digital data stream 122 by X bytes. And the digital data stream 122 is mapped to a (virtual concatenation) virtual container in a scrambled state.

FIG. 22 is a diagram illustrating an example of a device configuration according to an embodiment of the present invention, in which the length adjustment of the byte sequence of the subchannel is performed by the first format converter, and between the first format converter and the second format converter The digital data stream in which the subchannel is formed shows an operation in the case where the digital data stream is transferred in a descrambled state.
The data frames 311, 321, 341, 331, 332, 322 on the digital data signal 110 are converted into data frames 311, 321, 322 in order by attribute on the digital data stream 120 by the first format converter 510. , 331, 332, and 341, and by adjusting the length of the byte string of each subchannel by means 571 for adding idle data, subchannels 211, 212, 213, and 214 are placed on the digital data stream 121. It is formed as a byte sequence of X bytes.
The digital data stream 121 is descrambled by the descrambling function 572 and transferred from the first format converter 510 to the second format converter 520. The second format conversion device 520 maps the digital data stream 120 to a virtual container (virtual concatenation) in a scrambled state.

In the above two examples, since the digital data stream in the data processing apparatus is descrambled, it can be said that it is a suitable configuration to transfer using the electric signal interface. As an electrical signal interface, for example, SPI4-2 (System Packet Interface, Level 2, Phase 2), which is made IA by OIF, can be used.
This SPI4-2 is also adopted by the NPF (Network Processing Forum) as the signal interface standard NPSI (NPF Streaming Interface), and its high availability as a part and technology greatly contributes to the cost reduction of equipment. is there.

  When the digital data stream is 10 Gigabit Ethernet (registered trademark), the above scrambling and descrambling are performed only for the 64B payload of the 64B / 66B code.

FIG. 23 shows mapping of control bytes for staff operation using (virtually linked) virtual containers according to an embodiment of the present invention.
On the digital data stream 120, sub-channel byte sequences 211, 212, 213, and 214 are arranged. Each of these byte strings is X bytes in length. The digital data stream 120 in which the subchannel is formed is mapped to the container 420 as a byte string. Here, for each X byte corresponding to the subchannel byte string, a control byte 731 for SONET / SDH stuff operation, 732, 733, 734,... (Hereinafter not shown) are inserted.
At this time, the digital data stream 120 is mapped in a scrambled state. Container 420 further includes X virtual containers 430-1, 430-2, ..., 430-X plus virtual container 480 for mapping control bytes for staff operations (X + 1) virtual The container is mapped by byte disinterleave.
As a result, the byte strings of the subchannels 211, 212, 213, and 214 are distributed one byte at a time to the X virtual containers 430-1 to X, and the control bytes for staff operation 731, 732, 733, 734,. Mapping is concentrated on the virtual container 480. The subsequent handling of the X virtual containers 430-1 to X is the same as that described with reference to FIGS.

  The procedure for inserting the stuff operation control bytes into the digital data stream is performed via a container for convenience of explanation. However, it is not necessary to actually use the container, and the sequential processing as a byte sequence is performed. Realize it.

FIG. 24 shows the mapping of the LSS signal using the staff operation control byte according to the embodiment of the present invention in this example for 10 Gigabit Ethernet (registered trademark).
The 10-Gigabit Ethernet LSS signal is embedded utilizing a portion of the interframe gap, as shown in column 740 in column row 603, and the four LSS signal bytes 741, 742, 743 , 744. These LSS signal bytes are mapped to stuff information bytes (indicated by S) in a stuff operation control byte sequence 730 composed of stuff operation control bytes 731, 732, 733, 734.
As a result of each staff operation control byte being mapped to the virtual container 480 in the procedure as described above, the LSS signal bytes 741, 742, 743, 744 are the negative stuff information bytes 751, 752, 753, Each is mapped to 754.

Here, the negative staff operation using the control byte for staff operation will be described. In the stuff operation control byte sequence, the stuff operation control byte preceding one stuff information byte (S) includes a fixed stuff byte (R) and five stuff determination bytes (J).
The fixed stuff byte is a fixed pattern of “11111111”. The stuff determination byte is composed of a 7-bit fixed pattern and a 1-bit stuff determination bit C. When the stuff determination bits of the five stuff determination bytes are “CCCCC = 11111”, a negative stuff operation is performed, and information (ie, corresponding to the LSS signal byte) is entered in the stuff information byte.
On the other hand, when “CCCCC = 00000”, the stuff information byte contains a meaningless value. This stuff determination is performed by majority logic considering a bit error of 1 to 2 bits in C bits.

FIG. 25 shows mapping of control bytes for staff operation to (virtually linked) virtual containers corresponding to each sub-channel according to an embodiment of the present invention.
The procedure until the sub-channel byte sequences 211, 212, 213, and 214 on the digital data stream 120 are distributed and mapped to the X virtual containers 430-1 to X is the same as that described in FIG. . When mapping by virtual interleaving from the virtual containers 430-1 to X to containers 441, 442,... (Hereinafter not shown), stuff operation control bytes 761, 771,. The staff operation control bytes 761, 771,... Are used for staff operations related to (virtually connected) virtual containers 411, 412,.

  Note that the procedure for mapping the control bytes for staff operation to the above (virtually linked) virtual container is performed via the container for convenience of explanation, but it is not necessary to actually use the container, and the sequential byte sequence It may be realized by simple processing.

FIG. 26 illustrates subchannel identification by an input interface according to an embodiment of the present invention.
Data frames 311, 321 and 322, 331 and 332, 341 respectively input by the digital data signals 111, 112, 113, 114 are converted into digital data streams corresponding to the input interface by the first format converter 510. The subchannels 211, 212, 213, and 214 on 120 are arranged as byte strings forming each subchannel.

FIG. 27 illustrates subchannel identification by output interface according to an embodiment of the present invention.
The data frames 311, 321 and 322, 331 and 332, 341 input by the digital data signal 110 are respectively transmitted on the digital data stream 120 by the first format converter 510 corresponding to the output interface of each data frame. The subchannels 211, 212, 213, and 214 are arranged as byte strings forming each subchannel. The subchannels 211, 212, 213, and 214 are output as digital data streams 131, 132, 133, and 134, respectively, and are restored in such a way that the data frames 311, 321 and 322, 331, and 332 and 341 are included.

FIG. 28 illustrates subchannel identification by VLAN tag according to an embodiment of the present invention.
Data frames 311, 321, 341, 331, 332, 322 (Ethernet) are input by a digital data signal 110 (Ethernet). A data frame 311 (Ethernet (registered trademark)) includes a preamble / SFD • 811, a transmission destination MAC address field 812, a transmission source MAC address field 813, a Type / Length field 814, a data field 815, and an FCS field 816. A VLAN tag field 817 is inserted.
The first format conversion device 510 uses the VLAN tag or part thereof in the VLAN tag field 817 as an attribute of the data frame 311 (Ethernet (registered trademark)), and the digital data stream 120 (Ethernet (registered trademark)). The sub-channel 211 on which the data frame 311 (Ethernet (registered trademark)) is arranged is determined.

FIG. 29 illustrates identification of subchannels by MAC address according to an embodiment of the present invention.
Data frames 311, 321, 341, 331, 332, 322 (Ethernet) are input by a digital data signal 110 (Ethernet). The data frame 321 (Ethernet (registered trademark)) includes a preamble / SFD 821, a transmission destination MAC address field 822, a transmission source MAC address field 823, a Type / Length field 824, a data field 825, and an FCS field 826.
The first format conversion apparatus 510 uses the MAC address or part of the MAC address of the transmission destination MAC address field 822 and the transmission source MAC address field 823 as the attribute of the data frame 321 (Ethernet (registered trademark)), and (Ethernet A subchannel 212 is determined that arranges (Ethernet) data frames 321 on the (digital) digital data stream 120.

FIG. 30 illustrates subchannel identification by IP address according to an embodiment of the present invention.
Data frames 311, 321, 341, 331, 332, and 322 (in Ethernet (registered trademark) encapsulating IP packets) are input by the digital data signal 110. A data frame 331 (Ethernet (encapsulated trademark) encapsulating an IP packet) includes a preamble / SFD 831, a destination MAC address field 832, a source MAC address field 833, a Type / Length field 834, a data field 835, and an FCS It consists of field 836.
The data field 835 stores an IP packet 851. The first format converter 510 uses the IP address in the IP address field 852 of the IP packet 851 or a part thereof as an attribute of the data frame 331 (for Ethernet (registered trademark) encapsulating the IP packet), and (Ethernet ( A sub-channel 213 is determined which arranges data frames 331 (Ethernet® encapsulating IP packets) on a digital data stream 120 (registered trademark).

FIG. 31 illustrates subchannel identification by MPLS label according to an embodiment of the present invention.
Data frames 311, 321, 341, 331, 332, and 322 (Ethernet (registered trademark) encapsulating packets with MPLS labels) are inputted by a digital data signal 110 (Ethernet (registered trademark)). A data frame 341 (Ethernet (registered trademark) encapsulating a packet with an MPLS label) includes a preamble / SFD 841, a destination MAC address field 842, a source MAC address field 843, a Type / Length field 844, and a data field 845. , An FCS field 846, and a Shim header 848 is inserted.
The first format converter 510 uses the label of the label field 861 of the Shim header 848 or a part thereof as an attribute of the data frame 341 (for Ethernet (registered trademark) encapsulating an MPLS labeled packet), and (Ethernet ( A subchannel 214 is determined which arranges a data frame 341 (Ethernet) encapsulating an MPLS labeled packet on a digital data stream 120 (registered trademark).

FIG. 32 illustrates identification of subchannels by a combination of an input interface and other identifiers according to an embodiment of the present invention.
Data frames 311, 321 and 322, 331, 332, and 341 (Ethernet) are input by digital data signals 111, 112, 113, and 114 (Ethernet), respectively. The first format converter 510 causes the data frames 311 and 341 (Ethernet) to correspond to the input interface, respectively, with the subchannel 211 on the digital data stream 120 (Ethernet) and In 214, they are arranged as byte strings forming each subchannel.
Data frames 321, 322, 331, and 332 (Ethernet (registered trademark)) encapsulate the input interface and the VLAN tag, MAC address, and (IP packet) of each (Ethernet (registered trademark)) data frame. Subchannels 212, 215, 214 referring to the IP address (if MPLS-encapsulated packets are encapsulated) or the MPLS label (not shown) or part of them, respectively. , 216 are arranged as byte strings forming each subchannel.

  In the above example, the data frame (of Ethernet (registered trademark)) is selected according to the VLAN tag, MAC address, IP address, MPLS label (not shown), or part of them and the input interface. Although an example of attribute is described, any other combination may be used.

FIG. 33 shows a configuration example of the first format conversion device according to the embodiment of the present invention when a FIFO queue is provided.
Data frames 311, 321, 341, 331, 332 and 322 are input by the digital data signal 110. These data frames are stored in the FIFO queue provided for each sub-channel in the first format converter 510. Data frame 311 is in FIFO queue 581 of subchannel 211, data frames 321 and 322 are in FIFO queue 582 of subchannel 212, data frames 331 and 332 are in FIFO queue 583 of subchannel 213, and data frame 341 is Are stored in the FIFO queue 584 of the subchannel 214, and then arranged in the digital data stream 120 as a byte sequence of X bytes in order for each subchannel.

  Here, the reason why the queue for each subchannel provided in the first format conversion device is the FIFO queue is that the order of the data frames is not reversed in each subchannel.

FIG. 34 shows an example of the configuration of the first format conversion device according to the embodiment of the present invention when band setting is performed using a FIFO queue.
Here, the data frame of the input digital data signal and the digital data stream in which the subchannel is formed are not shown. The first format conversion device 510 is provided with FIFO queues 581, 582, 583, and 584 for each subchannel. These FIFO queues set the depth and input / output speed of each FIFO queue according to the bandwidth control conditions of each subchannel by means 591 for setting the depth and input / output speed of the FIFO queue.

FIG. 35 shows a configuration example when discarding non-priority data frames at the front stage of the FIFO queue in the first format conversion apparatus according to the embodiment of the present invention.
Here, the data frame of the input digital data signal and the digital data stream in which the subchannel is formed are not shown. The first format conversion device 510 is provided with FIFO queues 581, 582, 583, and 584 for each subchannel, and further, a means 592 for discarding non-priority data frames is provided in front of these FIFO queues. . The means 592 for discarding the non-priority data frame detects the non-priority data frame from the input data frame, and discards the non-priority data frame without inputting it into the FIFO queue.

FIG. 36 shows a configuration example when discarding a non-priority data frame at the latter stage of the FIFO queue in the first format conversion apparatus according to the embodiment of the present invention.
Here, the data frame of the input digital data signal and the digital data stream in which the subchannel is formed are not shown. The first format conversion device 510 is provided with FIFO queues 581, 582, 583, and 584 for each subchannel, and further, a means 593 for discarding non-priority data frames is provided at the subsequent stage of these FIFO queues. .
The means 593 for discarding the non-priority data frame detects the non-priority data frame from the data frame output from the FIFO queue, and discards the non-priority data frame without arranging it in the digital data stream.

FIG. 37 shows a configuration example in the case of performing priority control by discarding non-priority data frames in the first format conversion device according to the embodiment of the present invention.
Here, the data frame of the input digital data signal and the digital data stream in which the subchannel is formed are not shown. The first format conversion apparatus 510 is provided with FIFO queues 581, 582, 583, and 584 for each subchannel, and a means 592 for discarding non-priority digital frames is provided in the preceding stage of these FIFO queues. A non-priority digital frame identification function 594 is provided.
The non-priority digital frame identification function 594 identifies the priority of the digital frame with reference to the Type of Service field of the IP packet, the EXP bit of the Shim header, etc., and sets it in the means 592 for discarding the non-priority digital frame. . The means 592 for discarding the non-priority digital frame detects the non-priority data frame from the input digital frames, and discards the non-priority data frame without entering the FIFO queue.
In addition, here, the interaction between the means for discarding the non-priority data frame provided in the previous stage of the FIFO queue and the non-priority data frame identification function is illustrated, but the means for discarding the non-priority data frame is the FIFO queue. The same applies to the case where it is provided in the subsequent stage.

  The embodiments of the present invention have been described above with reference to the drawings. However, specific configurations are not limited to these embodiments, and design changes and the like within a scope not departing from the gist of the present invention are possible. Is possible.

  As described above, according to the data processing method and data processing apparatus of the present invention, virtual subchannels are formed in order on the digital data stream for each X byte, and this (unscrambled state) By mapping by byte disinterleave to X virtual containers and then by byte interleaving to (virtual concatenation) virtual containers, each byte of subchannel once distributed to X virtual containers (virtual concatenation) ) Since mapping can be concentrated on the virtual container, it has a remarkable effect of cutting through the sub-channel of the digital data stream to the SONET / SDH tributary.

  Therefore, when the configuration includes a SONET / SDH switch chip, it is possible to connect between the line correspondence unit and the switch fabric unit of a network device that performs packet switching, such as a router, via a SONET / SDH tributary. . Therefore, since the backplane interface in the router can be a standardized interface that does not include any proprietary format, it is possible to open the router form factor, which greatly contributes to cost reduction. is there.

Further, when the SONET / SDH transmission apparatus is provided, the virtual container to which the subchannel is mapped can be transferred on the SONET / SDH network configured by the SONET / SDH transmission apparatus. That is, packet-based digital data traffic such as Ethernet (registered trademark) can be transferred over the SONET / SDH network with flexible granularity.
When the SONET / SDH cross-connect is used, the virtual container to which the subchannel is mapped can be cross-connected (time slot exchange) with the SONET / SDH cross-connect. That is, it is possible to provide flexible granularity to packet-based digital data traffic such as Ethernet (registered trademark) and control the route on the SONET / SDH network while transferring it on the SONET / SDH network.

It is a figure which shows the concept of the data processing method (mapping) by embodiment of this invention. It is a figure which shows the concept of the data processing method (demapping) by this embodiment. It is a figure which shows the concept of the data processing method (demapping for every VC) by this embodiment. FIG. 4 is a diagram illustrating subchannel mapping from a digital data stream to a virtual container according to the present embodiment. FIG. 4 is a diagram illustrating demapping from a virtual container to a digital data stream according to the present embodiment. It is a figure which shows the demapping for every VC from a virtual container to a digital data stream by this embodiment. It is a figure which shows the principle of adjustment (length adjustment of the subchannel byte sequence) of encoding and byte multiplexing according to this embodiment, taking 10 gigabit Ethernet (registered trademark) as an example. It is a conceptual diagram which shows the basic composition of the data processor (mapping) by this embodiment. It is a conceptual diagram which shows the basic composition of the data processor (demapping) by this embodiment. It is a conceptual diagram which shows the basic composition of the data processor (demapping for every VC) by this embodiment. It is a figure which shows distribution of the output data stream in formation of the subchannel on the digital data stream of this embodiment. It is a figure which shows the example of an apparatus structure in case this embodiment contains a SONET / SDH switch chip. It is a figure which shows the apparatus structural example in the case of including a SONET / SDH apparatus of this embodiment. It is a figure which shows the apparatus structural example in the case of including SONET / SDH cross-connect of this embodiment. It is a figure which shows the operation | movement (byte-string length adjustment) regarding a synchronizing signal (each subchannel) by this embodiment. It is a figure which shows the operation | movement (byte sequence length adjustment and mapping head byte detection) regarding the synchronizing signal (each subchannel and every subchannel group) by this embodiment. It is a figure which shows the operation | movement (mapping head byte detection) regarding a synchronizing signal (per subchannel) by this embodiment. It is a figure which shows the operation | movement (mapping head byte detection) regarding a synchronizing signal (each subchannel group) by this embodiment. It is a figure which shows the operation | movement when the byte-string length of a subchannel is adjusted with the 2nd format conversion apparatus of the apparatus structural example of this embodiment, and the digital data stream inside a data processor is scrambled. It is a figure which shows the operation | movement when the byte sequence length of a subchannel is adjusted with the 1st format converter of the apparatus structural example of this embodiment, and the digital data stream inside a data processor is scrambled. It is a figure which shows operation | movement when the byte-string length of a subchannel is adjusted with the 2nd format conversion apparatus of the apparatus structural example of this embodiment, and the digital data stream inside a data processing apparatus is descrambled. It is a figure which shows the operation | movement when the byte sequence length of a subchannel is adjusted with the 1st format converter of the apparatus structural example of this embodiment, and the digital data stream inside a data processor is descrambled. It is a figure which shows the mapping of the control byte for staff operation using one virtual container with respect to the digital data stream by this embodiment. It is a figure which shows the mapping of the LSS signal using the control byte for staff operation | movement by this embodiment. It is a figure which shows the mapping of the control byte for staff operation to a subchannel (virtual container) by this embodiment. It is a figure which shows the identification of the subchannel by the input interface of this embodiment. It is a figure which shows the identification of the subchannel by the output interface of this embodiment. It is a figure which shows the identification of the subchannel by a VLAN tag of this embodiment. It is a figure which shows the identification of the subchannel by MAC address of this embodiment. It is a figure which shows the identification of the subchannel by an IP address of this embodiment. It is a figure which shows the identification of the subchannel by the MPLS label of this embodiment. It is a figure which shows the identification of the subchannel by the combination of an input interface and another identifier of this embodiment. It is a figure which shows the structural example at the time of providing the FIFO queue of the 1st format conversion apparatus of this embodiment. It is a figure which shows the structural example in the case of performing a bandwidth setting using the FIFO queue of the 1st format conversion apparatus of this embodiment. FIG. 6 is a diagram illustrating a configuration example when discarding a non-priority frame in the first stage of a FIFO queue in the first format conversion apparatus of the present embodiment. FIG. 6 is a diagram illustrating a configuration example when non-priority frame discarding is performed at the latter stage of the FIFO queue in the first format conversion device of the present embodiment. FIG. 5 is a diagram illustrating a configuration example in the case of performing priority control by discarding non-priority frames in the first format conversion device of the present embodiment.

Explanation of symbols

110-114, 120-123, 130-134, 141, 142 ... Digital data stream (signal).
211 to 216 ... Subchannel (and its byte sequence).
311, 321, 322, 331, 332, 341 ... Data frame.
401 to 405 ... SONET / SDH link (OC-192 / STM-64).
411 to 418 (virtual connection) Virtual container (VC-4-Nv).
420, 470 ... Container (C-4-Xc).
430-1 to X, 460-1 to X, 480 ... Virtual container (VC-4).
441, 442, 451, 452 ... Container (C-4-Nc).
501, 502: Data processing device.
510 ··· First format converter.
520 to 522 ... Second format conversion device.
530 to 532, 5301 to 5304, a third format conversion device.
550… SONET / SDH switch chip (time slot interchange).
551, 552 ... SONET / SDH equipment.
553, 554… SONET / SDH cross-connect.
561 ... Means for inserting a synchronization signal.
562 ... Means for extracting a synchronization signal.
571 ... Means for adding idle data (so that the byte sequence of the subchannel is X bytes).
572… Descramble function.
581-584 ... FIFO queue.
591 ・ ・ ・ ・ ・ ・ Means to set FIFO queue depth and I / O speed.
592, 593 ... Means for discarding non-priority data frames.
594 ... Non-priority data frame identification function.
601 to 603 ... Column row.
630-1 to p, 640-1 to q ... 64B / 66B codes.
631-1 ~ p, 641-1 ~ q ... 2-bit header.
632-1 ~ p, 642-1 ~ q ··· 64B payload.
610-1 ~ p, 620-1 ~ q ... 2 column block.
711 to 714 ... Synchronization signal (for each subchannel).
721 ... Sync signal (for each subchannel group).
730… Control byte sequence for staff operation.
731 to 734, 761, 771 ... Control bytes for staff operation.
740 ··· Column (LSS signal).
741 to 744 ... LSS signal byte.
751-754 ... Negative staff information byte.
811, 821, 831, 841 ... Preamble / SFD (for Ethernet frames).
812, 813, 822, 823, 832, 833, 842, 843 ... MAC address field (for Ethernet frames).
814, 824, 834, 844... Type / Length field (for an Ethernet frame).
815, 825, 835, 845 ··· Data field (for Ethernet frames).
816, 826, 836, 846 ... FCS field (of Ethernet frame).
817 ... VLAN tag field (for Ethernet frames).
848 ... Shim header field (for Ethernet frames).
851… IP packet.
852… IP address field (for IP packets).
861… Label field (in Shim header).

Claims (37)

A first step of inputting at least one digital data signal;
A second step of generating at least one digital data stream;
The data frame included in the digital data signal input in the first step is sequentially sorted according to the attribute of the data frame on the digital data stream generated in the second step according to the attribute of the data frame. A third step of arranging;
In the digital data stream in which the data frames are arranged in the third step, a digital data string including at least one of the data frames having the same attribute has a set length X (X is a positive value) A fourth step of adding idle data to be an integer) byte sequence;
A fifth step of mapping the digital data stream to which the idle data is added in the fourth step to X virtual containers by byte disinterleaving without being scrambled;
A sixth step of mapping each of the X virtual containers mapped in the fifth step to at least one virtual container by byte interleaving;
A data processing method characterized by comprising: A seventh step of encoding the digital data stream to which idle data is added in the fourth step;
2. The data according to claim 1, wherein an integer obtained by dividing an integer multiple of a least common multiple of the number of code bits of the encoding and the number of bits of the bytes by the number of bits of the bytes is used as the integer X. Processing method. An eighth step of mapping the at least one virtual container mapped in the fifth step to the X virtual containers by byte disinterleaving in order for each of the at least one virtual container;
A ninth step of mapping and outputting each of the X virtual containers mapped from the at least one virtual container in the eighth step to a digital data stream by byte interleaving;
The data processing method according to claim 1, further comprising:   3. The data according to claim 1, further comprising an eighth step of mapping and outputting each of the at least one virtual container mapped in the fifth step to a digital data stream. Processing method.   The method further comprises a tenth step of selecting the digital data stream in which the data frames are arranged from the at least one digital data stream generated in the second step according to the attribute of the data frame. 5. The data processing method according to claim 1, wherein the data processing method is characterized. An eleventh step of arranging the data frames in order for each of the attributes on the digital data stream generated in the second step and inserting a synchronization signal;
The synchronization signal is extracted from the digital data stream in which the synchronization signal has been inserted in the eleventh step, and the digital data sequence including the data frame having at least one of the same attributes is a byte sequence of X bytes A twelfth step of adding idle data to
6. The data processing method according to claim 1, further comprising: Idle data is added to the digital data stream generated in the second step so that at least one digital data sequence including the data frame having the same attribute is an X-byte byte sequence, and synchronization is performed. An eleventh step of inserting a signal;
A twelfth step of extracting the synchronization signal from the digital data stream into which the synchronization signal has been inserted in the eleventh step and mapping it to the X virtual containers;
6. The data processing method according to claim 1, further comprising:   8. The data processing method according to claim 6, wherein a synchronization signal is inserted for each of the attributes of the data frame in the eleventh step.   8. The data processing method according to claim 6, wherein a synchronization signal is inserted in each cycle of the attribute of the data frame in the eleventh step. A thirteenth step of inserting n (n is a positive integer) bytes of stuff operation control bytes for every X bytes of the digital data stream generated in the second step;
A fourteenth step of generating n virtual containers in addition to the X virtual containers to which the digital data stream is mapped in the fifth step to (X + n) virtual containers;
A fifteenth step of mapping the digital data stream in which the stuff operation control bytes are inserted in the thirteenth step to the (X + n) virtual containers generated in the fourteenth step by byte disinterleave;
A sixteenth step of mapping each of the X virtual containers mapped in the fourteenth step to at least one virtual container by byte interleaving;
A seventeenth step of mapping each of the n virtual containers generated in the fourteenth step to at least one virtual container by byte interleaving;
An eighteenth step of performing a negative stuff operation on the at least one virtual container mapped from the n virtual containers using the staff operation control byte inserted in the thirteenth step;
A nineteenth step of generating information bits by the negative stuffing operation of the eighteenth step;
A twentieth step of transmitting a link layer control signal using the information bits generated in the nineteenth step;
The data processing method according to claim 1, further comprising: When mapping the X virtual containers to which the digital data stream is mapped in the fifth step to the at least one virtual container in the sixth step by byte interleaving, the at least one virtual container A thirteenth step of inserting a control byte for staff operation in each of
A fourteenth step of performing a negative staff operation on at least one virtual container of at least one virtual container in which the staff operation control byte is inserted in the thirteenth step;
A fifteenth step of generating an information bit by the negative stuff operation of the fourteenth step;
Using the information bits generated in the fifteenth step, a sixteenth step of transmitting a link layer control signal;
The data processing method according to claim 1, further comprising:   12. The method according to claim 1, further comprising a 21st step of setting an input interface number of the digital data signal to which the data frame is input as the attribute of the data frame. Data processing method.   12. The method according to claim 1, further comprising a 21st step of setting an output interface number of the digital data stream to which the data frame is output as the attribute of the data frame. Data processing method.   2. The method according to claim 1, further comprising a 21st step of setting an Ethernet (registered trademark) frame VLAN tag in the input data frame or a part of the VLAN tag as the attribute of the data frame. 11. The data processing method according to any one of items 11.   2. The method according to claim 1, further comprising a 21st step of setting the MAC address of the Ethernet (registered trademark) frame in the input data frame or a part of the MAC address as the attribute of the data frame. 11. The data processing method according to any one of items 11.   12. The method according to claim 1, further comprising a 21st step of setting an IP address of the IP packet in the input data frame or a part of the IP address as the attribute of the data frame. The data processing method described in the section.   12. The method according to claim 1, further comprising a twenty-first step in which a Shim header label or a part of the Shim header label in the input data frame is used as the attribute of the data frame. A data processing method according to any of the above sections.   The input interface number of the digital data signal to which the data frame is input, the output interface number of the digital data stream to which the data frame is output, and the VLAN of the Ethernet (registered trademark) frame in the input data frame Tag or part of the VLAN tag, the MAC address of the Ethernet (registered trademark) frame in the input data frame or a part of the MAC address, and the IP address of the IP packet in the input data frame or A combination of at least two of the part of the IP address and the label of the Shim header or the part of the Shim header in the input data frame is used as the attribute of the data frame. 12. The data processing according to claim 1, further comprising a step. Method. In a data processing device comprising a first format conversion device and a second format conversion device,
The first format conversion device,
Digital data signal input means for inputting at least one digital data signal;
First digital data stream generating means for generating at least one digital data stream;
Data frames in the digital data signal input from the digital data signal input means are arranged on the digital data stream generated by the first digital data stream generation means in order according to the attribute of the data frame. Having data frame arrangement means;
The second format conversion device,
First idle data to which idle data is added so that a digital data sequence including the data frame having at least one of the same attributes is a byte sequence having a set length X (X is a positive integer) bytes Additional means;
A first mapping means for mapping the digital data stream in which the data frame arranging means has arranged the data frames to X virtual containers by byte disinterleaving without being scrambled;
Each of the X virtual containers mapped by the first mapping means has second mapping means for mapping to at least one virtual container by byte interleaving;
A data processing apparatus. The data frame arrangement means further comprises a digital data stream encoding means for encoding a digital data stream in which the data frames are arranged;
As the integer X, an integer obtained by dividing an integer multiple of the least common multiple of the number of code bits encoded by the digital data stream encoding means and the number of bits of the bytes by the number of bits of the bytes is used. 20. A data processing apparatus according to claim 19. A third mapping means for mapping at least one virtual container mapped by the second mapping means, in order for each of the at least one virtual container, to byte X virtual containers by byte disinterleaving;
A fourth mapping means for mapping each of the X virtual containers mapped by the third mapping means to a digital data stream by byte interleaving;
21. The data processing apparatus according to claim 19, further comprising a third format conversion apparatus having:   20. The third format conversion device further comprising a third format conversion device having third mapping means for mapping at least one virtual container mapped by the second mapping means to a digital data stream. 20. The data processing device according to 20.   23. The data processing device according to claim 21, further comprising at least one SONET / SDH switch chip that connects the second format conversion device and the third format conversion device.   23. The data processing device according to claim 21, further comprising at least one SONET / SDH device that connects the second format conversion device and the third format conversion device.   25. The data processing apparatus according to claim 24, wherein the at least one SONET / SDH apparatus further includes at least one SONET / SDH cross-connect apparatus. The second format conversion device adds idle data to the digital data stream so that a digital data sequence including the data frame having the same attribute at least one is an X-byte byte sequence. Idle data addition means;
A fifth mapping means for mapping the digital data stream with the idle data added by the second idle data adding means to the X virtual containers by byte disinterleaving in a non-scrambled state;
The data processing device according to any one of claims 19 to 25, further comprising: The second format conversion device adds idle data to the digital data stream so that a digital data sequence including the data frame having at least one of the same attributes is an X-byte byte sequence. Further comprising idle data adding means,
The second format conversion device maps the digital data stream with the idle data added by the second idle data adding means to the X virtual containers by byte disinterleaving without being scrambled. Further comprising 5 mapping means,
The data processing apparatus according to any one of claims 19 to 25, wherein: The first format conversion device further includes second digital data stream generation means for generating the digital data stream without being scrambled;
The second format conversion device includes a digital data stream including at least one data frame having the same attribute as the digital data stream generated by the second digital data stream generation means. A second idle data adding means for adding idle data so that
The data processing apparatus according to any one of claims 19 to 25, wherein: The first format conversion device, the second digital data stream generation means for generating the digital data stream without being scrambled;
Idle data is added to the digital data stream generated by the second digital data stream generation means so that a digital data sequence including the data frame having the same attribute is at least one byte sequence. Second idle data adding means for
The data processing device according to any one of claims 19 to 25, further comprising: The first format conversion device further includes synchronization signal insertion means for inserting a synchronization signal into the digital data stream;
The second format conversion device further includes synchronization signal extraction means for extracting the synchronization signal;
30. The data processing device according to claim 19, wherein the data processing device is a data processing device.   31. The data processing apparatus according to claim 30, wherein the synchronization signal insertion unit inserts the synchronization signal into the digital data stream for each attribute of the data frame.   31. The data processing apparatus according to claim 30, wherein the synchronization signal insertion unit inserts the synchronization signal into the digital data stream for each cycle of the attribute of the data frame.   The first format conversion device further includes a FIFO queue for storing the data frame in the digital data signal input from the digital data signal input means for each attribute of the data frame. 33. A data processing device according to any one of claims 19 to 32.   34. The data processing device according to claim 33, wherein the first format conversion device further comprises FIFO queue setting means for setting the depth and input / output speed of the FIFO queue for each attribute. Non-priority data frame detection means for detecting a non-priority data frame among the data frames in the digital data signal input from the digital data signal input means by the first format conversion device;
Non-priority data frame discarding means for discarding the non-priority data frame without storing it in the FIFO queue;
35. The data processing apparatus according to claim 33 or 34, further comprising: Non-priority data frame detection means for detecting a non-priority data frame among the data frames in the digital data signal input from the digital data signal input means by the first format conversion device;
Non-priority data frame discarding means for outputting the non-priority data frame from the FIFO queue and immediately discarding it,
35. The data processing apparatus according to claim 33 or 34, further comprising: The non-priority data frame detection means further comprises non-priority data frame identification means for identifying the non-priority data frame from information mounted on an IP header or Shim header of an IP packet in the data frame. 37. A data processing apparatus according to claim 35 or 36.

Images