Method for realizing ultra-high-speed bandwidth based on universal backplane
By extending and redefining the VPX interface of the 3U VPX chassis, using Aurroax8, Aurroax4 and PCIE3.0 signal definitions, combined with PCIE switching module, the problem of low utilization of VPX interface is solved, and the interconnection and update of higher bandwidth is achieved Flexible module combination to meet different application needs.
Patent Information
- Application Number
- CN202411201179.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-29
AI Technical Summary
Due to the limited number of VPX connector interfaces in the existing 3U VPX chassis, the interconnection relationship between the boards and cards cannot reach a large bandwidth when the device functional modules increase, and the bandwidth requirements of different applications are inconsistent, resulting in low utilization of VPX interface definition. .
By extending the VPX high-speed interface definition of playback module, acquisition module and storage module, using high-speed data interface defined by Aurroax8 and Aurroax4 signals, combined with high-speed serial interface defined by PCIE3.0 signals, PCIE switching module is added to configure different high-speed chains The number of channels can improve interface utilization and allow the playback module and acquisition module to be inserted and interchangeable, achieving a combination of multiple functions.
It improves the utilization rate of VPX interface definition, achieves higher bandwidth interconnection, enhances the universality and flexibility of the system, and can flexibly select module combinations under different bandwidth requirements to adapt to a variety of application scenarios.
Smart Images

Figure CN118708527B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication technology, and in particular to a method for realizing ultra-high-speed bandwidth based on a universal backplane. Background Art
[0002] As a universal chassis, the 3U VPX chassis has a limited number of VPX connector interfaces, which results in the interconnection between boards not being able to achieve a higher bandwidth when the number of device function modules in the chassis increases. At the same time, different applications have different bandwidth requirements. However, by constraining the VPX definition of conventional modules, the backplane definition can be universally designed to be compatible with higher-speed bandwidth.
[0003] In the past, signal processing modules were usually designed based on the bandwidth requirements of actual applications:
[0004] 1. Use as few interfaces as possible to implement functions to avoid redundant design, resulting in low utilization of VPX interface definitions;
[0005] 2. Due to the diversity of applications, previous modules cannot be compatible when implementing highly similar functions. Summary of the invention
[0006] The purpose of the present invention is to provide an ultra-high-speed bandwidth implementation method based on a universal backplane, expand the VPX high-speed interface definition of an acquisition module, a playback module and a storage module, and improve the utilization rate of the VPX interface definition.
[0007] To achieve the above objectives, the solution of this application is:
[0008] The present invention provides an ultra-high-speed bandwidth implementation method based on a universal backplane, including a playback module, an acquisition module and a storage module, wherein the VPX high-speed interfaces of the acquisition module and the playback module both include a high-speed data interface defined by an Aurroax8 signal and an extended high-speed interface defined by an Aurroax4 signal;
[0009] The extended high-speed interface of the acquisition module and the extended high-speed interface of the playback module are connected via the Aurroax4 signal line;
[0010] The VPX high-speed interface of the storage module includes two groups of high-speed data interfaces and extended data interfaces both of which are defined by Aurroax8 signals. The high-speed data interface of the storage module is connected to the high-speed data interface of the acquisition module through the Aurroax8 signal line, and the extended data interface of the storage module is connected to the high-speed data interface of the playback module through the Aurroax8 signal line.
[0011] Specifically, the universal backplane also includes a PCIE switching module, the VPX high-speed interface of the PCIE switching module includes a high-speed serial interface defined by PCIE3.0 signals, and the VPX high-speed interfaces of the playback module, the acquisition module and the storage module also include a high-speed serial interface defined by PCIE3.0 signals. The high-speed serial interfaces of the playback module, the acquisition module and the storage module are respectively connected to the high-speed serial interface of the PCIE switching module through PCIE signal lines, so that data transmission and communication connections are carried out between the modules through the PCIE switching module.
[0012] Specifically, the high-speed serial interface of the PCIE switching module can be configured to have different numbers of high-speed links according to the needs of signal bandwidth to increase the total bandwidth.
[0013] Specifically, the high-speed serial interface of the PCIE switching module is configured as a PCIE3.0x4 high-speed serial interface defined by PCIE3.0x4 signals; the high-speed serial interface of the playback module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals, and the high-speed serial interface of the storage module is configured as a PCIE3.0x4 high-speed serial interface defined by PCIE3.0x4 signals. The PCIE3.0x4 high-speed serial interface of the PCIE switching module is respectively connected to the PCIE3.0x8 high-speed serial interface of the playback module and the PCIE3.0x4 high-speed serial interface of the storage module.
[0014] Specifically, the high-speed serial interface of the PCIE switching module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals; the high-speed serial interface of the acquisition module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals, and the PCIE3.0x8 high-speed serial interface of the PCIE switching module is connected to the PCIE3.0x8 high-speed serial interface of the acquisition module.
[0015] Specifically, the acquisition module and the playback module are respectively separately configured with a DSP1 module and a DSP2 module for assisting in data processing. The VPX high-speed interfaces of the acquisition module and the playback module also include a data interconnection interface defined by SRIO signals; the VPX high-speed interfaces of the DSP1 module and the DSP2 module both include a data interconnection interface defined by SRIO signals, the data interconnection interface of the acquisition module is connected to the data interconnection interface of the DSP1 module via an SRIOx4 signal line, and the data interconnection interface of the playback module is connected to the data interconnection interface of the DSP2 module via an SRIOx4 signal line.
[0016] Specifically, the boards inserted into the playback module and the acquisition module are interchangeable. The playback module can be used to insert a playback board or an acquisition board, and the acquisition module can be used to insert a playback board or an acquisition board. In this way, two playback boards or two acquisition boards can exist in the system at the same time, which can realize a single acquisition or a single playback function, or a combination of single acquisition and single playback, dual acquisition, or dual playback functions.
[0017] Specifically, the VPX high-speed interfaces of the DSP1 module and the DSP2 module also include high-speed serial interfaces defined by PCIE3.0 signals. The high-speed serial interfaces of the DSP1 module and the DSP2 module are respectively connected to the high-speed serial interface of the PCIE switching module through PCIE signal lines.
[0018] Specifically, the universal backplane also includes a CPU module, the VPX interface of the CPU module includes a control interface defined by PCIE2.0 signals, and the VPX interfaces of the playback module, acquisition module, DSP1 module and DSP2 module all include control interfaces defined by PCIE2.0 signals. The control interface of the CPU module is respectively connected to the control interfaces of the playback module, acquisition module, DSP1 module and DSP2 module through PCIE2.0 signal lines.
[0019] Specifically, the VPX interfaces of the acquisition module, playback module, storage module, DSP1 module and DSP2 module all include LVDS differential signal interfaces, and the differential signal interfaces between the acquisition module, playback module, storage module, DSP1 module and DSP2 modules are respectively connected to each other through LVDS signal lines.
[0020] The inventive concept of the present invention is:
[0021] In the prior art, due to the limitation of the number of VPX connector interfaces, when the number of functional modules of the equipment in the chassis increases, the interconnection relationship between the boards is limited in disguise and cannot achieve a larger bandwidth. At the same time, the bandwidth requirements for different applications are inconsistent. However, it is still possible to define a universal design for the backplane to be compatible with higher-speed bandwidth by constraining the VPX definition of conventional modules. However, in the design, the whole machine is usually designed according to the bandwidth requirements in the actual application, so that the functions are implemented with the most frugal interfaces to avoid redundant designs, resulting in low utilization of VPX interface definitions.
[0022] In order to solve the problem of low utilization rate of VPX definition, the present application expands the VPX high-speed interface definition of playback module, acquisition module and storage module; adds a group of Aurroax4 signals to the acquisition module as an extended high-speed interface; considering that a more universal design includes acquisition module and storage module, a group of playback modules is added to realize a universal chassis while expanding a group of Aurroax8 of storage module. Under this architecture, the function of a single acquisition or a single playback module can be realized, and the function combination of single acquisition and single playback module, dual acquisition module or dual playback module can also be realized at the same time. Based on the backplane expansion interface and standardized definition, there is no need to redesign, and only modules need to be selectively installed according to needs.
[0023] The beneficial effects of the present invention are:
[0024] This application improves the problem of low utilization of VPX definitions by expanding the VPX high-speed interface definitions of the playback module, acquisition module, and storage module. At the same time, the boards inserted into the playback module and the acquisition module are interchangeable. The playback module can be used to insert a playback board or an acquisition board, and the acquisition module can be used to insert a playback board or an acquisition board. In this way, two playback boards or two acquisition boards can exist in the system at the same time, which can realize a single acquisition or a single playback function, or a combination of single acquisition and single playback, dual acquisition, or dual playback functions. Improve the versatility of the chassis. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 A schematic diagram of a conventional acquisition storage chassis function VPX definition in the prior art;
[0026] Figure 2 A schematic diagram of the VPX definition of the acquisition, playback, processing and storage chassis function under the universal backplane provided in an embodiment of the present invention;
[0027] Figure 3 It is a schematic diagram of the LVDS full interconnection of the interface between the hybrid modules;
[0028] Figure 4 A schematic diagram of the interconnection relationship between the IO expansion module and the hybrid module provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0030] The relative arrangement of components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.
[0031] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.
[0032] Additionally, descriptions of well-known structures, functions, and configurations may be omitted for clarity and conciseness.One of ordinary skill in the art will recognize that various changes and modifications may be made to the examples described herein without departing from the spirit and scope of the present disclosure.
[0033] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.
[0034] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.
[0035] like Figure 1 As shown, the existing acquisition storage chassis includes a CPU module, an acquisition module, a DSP module, a storage module, and a radio frequency module matching the acquisition module. The function VPX is defined as:
[0036] The VPX interface definition of the CPU module includes two groups of PCIEx8 interfaces and SGMII interfaces, the storage module includes one group of Aurorax8 interfaces, one group of Aurorax4 interfaces and two groups of PCIEx4 interfaces, the acquisition module includes one group of Aurorax8 interfaces, two groups of PCIEx8 interfaces, one group of SRIOx4 interfaces, one group of PCIEx2 interfaces, one group of LVDSx2 interfaces, the DSP module includes one group of SRIOx4 interfaces, one group of PCIEx2 interfaces, one group of LVDSx2 interfaces, the RF module is connected to the acquisition module through the SPI bus, and the DSP module is connected to the CPU module through the SGMII signal line network;
[0037] The SRIOx4 interface of the DSP module is connected to the SRIO interface of the acquisition module through the SRIO signal line, the PCIEx2 interface of the DSP module is connected to the PCIEx2 interface of the acquisition module through the PCIE signal line, and the LVDSx2 interface of the DSP module is connected to the LVDSx2 interface of the acquisition module through the LVDS signal line.
[0038] The Aurorax8 interface of the acquisition module is connected to the Aurorax8 interface of the storage module through the Aurora signal line, the PCIEx8 interface of the acquisition module is connected to a group of PCIEx8 interfaces of the CPU module through the PCIE signal line, and another PCIEx8 interface of the CPU module is connected to the PCIEx4 interface of the storage module through the PCIE signal line.
[0039] It can be seen that Figure 1 In the VPX interface definition shown, in order to meet the bandwidth requirements, the interface implementation function will be implemented as frugally as possible to avoid redundant design, which will result in low utilization of the VPX interface definition. In this regard, the solution adopted by this application is as follows:
[0040] Example 1
[0041] like Figure 2 As shown, this embodiment provides an ultra-high-speed bandwidth implementation method based on a universal backplane. After the functional boards are inserted into the card slots of the universal backplane, a playback module, a capture module, a CPU module and a storage module are formed. The capture module and the playback module are separately configured with a DSP1 module and a DSP2 module for assisting data processing.
[0042] The VPX high-speed interfaces of the acquisition module and the playback module include a high-speed data interface defined by Aurroax8 signals, an extended high-speed interface defined by Aurroax4 signals, a data interconnection interface defined by SRIO signals, two pairs of PCIE2.0x1 signal interfaces, three pairs of LVDSx1 signal interfaces, and a group of SPI interfaces. The acquisition module includes a group of PCIE3.0x8 high-speed serial interfaces and a pair of SGMII signal interfaces defined by PCIE3.0 signals; the playback module uses a group of PCIE3.0x4 high-speed serial interfaces defined by PCIE3.0 signals;
[0043] It can be seen that the functions of the acquisition module and the playback module are inconsistent, but the definitions on the VPX slot are consistent, including a group of Aurorax8 and storage board interconnection to store the acquired data or store the data that needs to be played back; a group of expanded Aurorax4 is to strengthen the communication and data exchange between the two modules, which is also a disguised increase in the overall bandwidth; a group of SRIOx4 connected to the DSP module is an expansion of the FPGA signal processing capability; a group of PCIE3.0x4 is mainly to upload the processed data to the CPU through the switch board; it should be noted that the high-speed serial interface of the acquisition module is configured as PCIE3.0x8, and the high-speed serial interface of the playback module is also configured as PCIE3.0x8, but because the switch module has no extra interfaces, only x4 is connected in the connection relationship. In order to achieve interchange with the acquisition module, the actual playback module is configured as x8. Any other modules connected to the switch module interface can be interconnected through the switch module; the two groups of X8 of the CPU are connected to the switch board and the GPU module respectively; the GPU module can be used to expand the algorithm function processing of the CPU;
[0044] The VPX high-speed interface of the storage module includes two sets of high-speed data interfaces and extended data interfaces both defined by Aurroax8 signals, one set of PCIE3.0x4 high-speed serial interfaces defined by PCIE3.0 signals, a pair of SGMII signal interfaces, and three pairs of LVDSx1 signal interfaces;
[0045] The VPX interface of the CPU module includes four pairs of PCIE2.0x1 signal control interfaces defined by PCIE2.0 signals, two sets of PCIE3.0x8 high-speed serial interfaces and four pairs of SGMII signal interfaces;
[0046] The VPX high-speed interfaces of the DSP1 module and the DSP2 module also include a set of SRIOx4 data interconnect interfaces defined by SRIO signals, a set of PCIE3.0x4 high-speed serial interfaces, two sets of PCIE2.0x1 control interfaces, a pair of SGMII signal interfaces, and three pairs of LVDSx1 signal interfaces;
[0047] The interface connection between each module is as follows:
[0048] The extended high-speed interface of the acquisition module and the extended high-speed interface of the playback module are connected through the Aurroax4 signal line; the high-speed data interface of the storage module and the high-speed data interface of the acquisition module are connected through the Aurroax8 signal line, and the extended data interface of the storage module and the high-speed data interface of the playback module are connected through the Aurroax8 signal line. The data interconnection interface of the acquisition module is connected to the data interconnection interface of the DSP1 module through the SRIOx4 signal line, and the data interconnection interface of the playback module is connected to the data interconnection interface of the DSP2 module through the SRIOx4 signal line.
[0049] The four pairs of PCIE2.0x1 signal control interfaces of the CPU module are connected to a pair of PCIE2.0x1 signal interfaces of the playback module, acquisition module, DSP1 module and DSP2 module through PCIE2.0 signal lines, respectively, to realize the control of the functional boards inserted by the inserted CPU board for each module, and the four pairs of SGMII signal interfaces are connected to a pair of SGMII signal interfaces of the acquisition module, storage module, DSP1 module and DSP2 module through SGMII signal lines. The LVDS differential signal interfaces between the acquisition module, playback module, storage module, DSP1 module and DSP2 module are connected to each other through LVDS signal lines.
[0050] Specifically, by using the above-mentioned VPX signal definition and signal interconnection, the boards inserted into the playback module and the acquisition module are interchangeable, the playback module can be used to insert a playback board or an acquisition board, and the acquisition module can be used to insert a playback board or an acquisition board. In this way, two playback boards or two acquisition boards can exist in the system at the same time, which can realize a single acquisition or a single playback function, or a combination of single acquisition and single playback, dual acquisition or dual playback functions.
[0051] It is understandable that in order to solve the low utilization rate of VPX definition, the VPX high-speed interface definition has been expanded in the above-mentioned chassis VPX interface interconnection relationship; for example, the acquisition module has added a group of Aurroax4 signals as an extended high-speed interface; considering that a more general design includes acquisition modules and storage modules, a group of playback modules is also added to realize a general chassis while expanding a group of Aurroax8 of the storage module. Under this architecture, the functions of a single acquisition or a single playback module can be realized, and a single acquisition and single playback module, a dual acquisition module or a dual playback module function combination can also be realized at the same time. Based on the backplane expansion interface and standardized definition, there is no need for redesign, and only modules need to be selectively installed according to needs.
[0052] In order to further improve the utilization rate of the VPX high-speed interface, the present embodiment also includes a PCIE switching module for inserting a PCIE switching card. By extending the PCIE interface on the PCIE switching module, the utilization rate of the VPX high-speed interface is further improved while also increasing the number of PCIE high-speed links to increase the total bandwidth; the PCIE on each group of modules can communicate or transmit data with the remaining modules through the switching card. With the support of the universal backplane, when applied in a scenario with a smaller bandwidth, it only needs to be replaced with a small bandwidth module for acquisition / playback, and adapted by configuring different link numbers on the PCIE switch (such as x4 to x2, x8 to x4); similarly, when applied in a scenario with a large bandwidth, it only needs to be replaced with a large bandwidth module for acquisition / playback, and the PCIE is configured to the highest bandwidth link number to complete the conversion of ultra-high-speed bandwidth.
[0053] Specifically, the VPX high-speed interface of the PCIE switching module includes a high-speed serial interface defined by PCIE3.0 signals, and the VPX high-speed interfaces of the playback module, the acquisition module and the storage module also include high-speed serial interfaces defined by PCIE3.0 signals. The high-speed serial interfaces of the playback module, the acquisition module and the storage module are respectively connected to the high-speed serial interface of the PCIE switching module through PCIE signal lines, so that data transmission and communication connections are carried out between the modules through the PCIE switching module.
[0054] In order to increase the total bandwidth by configuring different high-speed link numbers according to the needs of signal bandwidth. The high-speed serial interface of the PCIE switching module is configured as four groups of PCIE3.0x4 high-speed serial interfaces defined by PCIE3.0x4 signals; the high-speed serial interface of the playback module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals and a group of PCIE3.0x8 high-speed serial interfaces defined by PCIE3.0x8 signals; the high-speed serial interface of the storage module is configured with a group of PCIE3.0x4 high-speed serial interfaces defined by PCIE3.0x4 signals; the high-speed serial interface of the acquisition module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals; the four groups of PCIE3.0x4 high-speed serial interfaces of the PCIE switching module are respectively connected to the PCIE3.0x4 high-speed serial interfaces of the playback module, the storage module, the DSP1 module and the DSP2 module. A group of PCIE3.0x8 high-speed serial interfaces of the PCIE switching module is connected to the PCIE3.0x8 high-speed serial interface of the acquisition module.
[0055] In order to improve the processing power of the CPU, the universal backplane in this embodiment also includes a GPU module. The VPX interface of the GPU module includes a group of PCIE3.0x8 high-speed serial interfaces. The two groups of PCIE3.0x8 high-speed serial interfaces of the CPU module are respectively connected to the PCIE3.0x8 high-speed serial interface of the GPU module and the PCIE3.0x8 high-speed serial interface of the PCIE switch module.
[0056] In order to increase the scalability of the universal backplane, an IO expansion module is also included. The IO expansion module is connected to the acquisition module, DSP1 module, storage module, playback module, DSP2 module and RF module through the SPI control bus. Specifically, Figure 4 As shown, the up-conversion module and the down-conversion module are connected to the IO extension module through the SPI bus; the DSP module is connected to the IO extension module through 1 SPI bus, 1 I2C bus, 1 GPI bus, 1 GPO bus and 2 LVDS signal lines; the playback module is connected to the IO extension module through 1 SPI bus, 1 I2C bus, 1 GPI bus, 1 GPO bus and 2 LVDS signal lines; the storage module is connected to the IO extension module through 1 SPI bus, 1 I2C bus, 1 GPI bus, 1 GPO bus and 2 LVDS signal lines; the acquisition module is connected to the IO extension module through 1 SPI bus, 1 I2C bus, 1 GPI bus, 1 GPO bus and 2 LVDS signal lines. All modules are connected to the IO expansion module, so that the SPI control signals of all inserted boards are interconnected to the IO expansion card, and then the IO expansion card selects which slot to interconnect the RF module. The RF module includes an up-conversion module and a down-conversion module. Since each acquisition module needs to be connected to a corresponding down-conversion module, each playback board is connected to an up-conversion module. Therefore, when the acquisition card and playback card are inserted into the hybrid module at will, the connection between the playback module and the acquisition module and the RF module can be flexibly configured through the IO expansion card. All modules are equipped with an IPMC management module and are interconnected.
[0057] The design method of the universal backplane in this embodiment is as follows:
[0058] The fully covered chassis design is compatible with the complete acquisition and playback system: roughly including
[0059] (a) Acquisition card + DSP co-processing + down-conversion RF module;
[0060] (b) Playback card + DSP co-processing + up-conversion RF module;
[0061] (c) CPU+GPU high-speed data stream processing and software algorithm implementation;
[0062] Since the high-speed interface definition of the acquisition and playback cards is symmetrical in the backplane design, the playback module can be configured as an acquisition module to become a dual acquisition to expand the acquisition bandwidth, or the acquisition module can be configured as a playback module to become a dual playback to expand the playback bandwidth. The modules can be disassembled and combined according to actual requirements, which has variability.
[0063] Based on the universal backplane, this application expands the chassis bandwidth in the following ways:
[0064] Under the full-coverage universal backplane, the complete system must be included, so the above modules are indispensable; and the bandwidth can be broadened in the following ways:
[0065] (a) Dual acquisition or dual playback can be added to expand the acquisition or playback bandwidth;
[0066] (b) Through the PCIE switch board, high-speed communication can be carried out between the modules and the PCIE3.0 interface can be provided to achieve high utilization of the interface. Otherwise, only two-way interconnection between two modules will limit the expansion of the backplane slots. For example, if the VPX interface is only connected to PCIEx8, a single board can only be connected to 4 other modules. By configuring the number of PCIE links, the bandwidth required by each can be adapted, so as to achieve comprehensiveness without excessive redundancy.
[0067] From a holistic perspective, the signal interconnection of the entire backplane can be divided into: three data layers connected by PCIE3.0, Aurora, and SRIO signal lines, a control layer connected by PCIE2.0 signal lines, a network layer using 1000BASE-X of the SGMII protocol, a control layer using the SPI signal bus, a chassis management layer, and a power supply layer. Specifically, when designing the connection of each signal line on the backplane and the definition of the VPX interface signal, ensure that the PCIE uplink channel routing of the computing module (playback module, storage module, and acquisition module) interconnecting the switching module is as short as possible; and minimize the crossing of signal routing on the backplane. Based on this, the interface position setting and signal definition of each VPX connector of each module are:
[0068] like Figure 2-Figure 3 As shown, Figure 3 The differential signal LVDS interconnection relationship between the modules is shown in FIG. 1 . Each module includes a semi-differential module VPX connector P0. The interface of the semi-differential module VPX connector P0 is defined as being used to connect a power supply to supply power to the inserted board.
[0069] The interface signal definitions of the DSP1 module and the DSP2 module are the same, and both include two fully differential module VPX connectors P1 and P2, and the interfaces of the connector P1 are divided into four groups of x4 transceiver modules 0, 1, 2, and 3, and the interface signals of the second group of x4 transceiver modules are defined as SRIOx4;
[0070] The interfaces of connector P2 are divided into four groups of x4 transceiver modules 0, 1, 2, and 3. Each group of x4 transceiver modules is divided into four pairs of differential signal interfaces. The interface signal of the 0th group of x4 transceiver modules is defined as: PCIE3.0x4. For the four pairs of differential signal interfaces A, B, C, and D of the 2nd group of x4 transceiver modules, the Ath pair of differential signal interfaces is defined as: PCIE2.0x1, and the Dth pair of differential signal interfaces is defined as: SGMII; for the four pairs of differential signal interfaces of the 3rd group of x4 transceiver modules, the Ath pair, the Bth pair, and the Cth pair of differential signal interfaces are all defined as: LVDS.
[0071] Specifically, the acquisition module also includes two fully differential module VPX connectors P1 and P2, and the interfaces of the connector P1 are divided into two groups of x4 transceiver modules 2 and 3 and one group of x8 transceiver modules 0 / 1, and the interface signal of the x8 transceiver module is defined as Aurorax8; the interface signals of the second group of x4 transceiver modules are defined as: SRIOx4, and the third group of x4 transceiver modules is defined as Aurorax4; the interfaces of the connector P2 are divided into two groups of x4 transceiver modules 2 and 3, and one group of x8 transceiver modules 0 / 1, and the interface signal of the x8 transceiver module is defined as PCIE3.0x8; the four pairs of differential signal interfaces A, B, C, and D of the second group of x4 transceiver modules, the differential signal interfaces of the A pair and the B pair are defined as: PCIE2.0x1, and the D pair of differential signal interfaces is defined as: SGMII;
[0072] The four pairs of differential signal interfaces A, B, C, and D of the third group of x4 transceiver modules define the differential signal interfaces of pairs A, B, and C as LVDS, and the differential signal interface of pair D as SPI.
[0073] The playback module also includes two fully differential module VPX connectors P1 and P2, which divide the interfaces of connector P1 into two groups of x4 transceiver modules 2 and 3 and one group of x8 transceiver modules 0 / 1, and define the interface signal of x8 transceiver module as Aurorax8; the interface signal of the second group of x4 transceiver modules is defined as SRIOx4, and the third group of x4 transceiver modules is defined as Aurorax4; it can be seen that the P1 connector interface definition of the acquisition module is consistent with that of the playback module, so the inserted acquisition board and playback board can be interchanged;
[0074] The interfaces of the connector P2 of the playback module are divided into four groups of x4 transceiver modules 0, 1, 2, and 3, and the interface signals of the 0th group of x4 transceiver modules are defined as PCIE3.0x4; the four pairs of differential signal interfaces A, B, C, and D of the 2nd group of x4 transceiver modules, the A and B pairs of differential signal interfaces are defined as: PCIE2.0x1; the four pairs of differential signal interfaces of the 3rd group of x4 transceiver modules, the A, B, and C pairs of differential signal interfaces are defined as: LVDS, and the D pair of differential signal interfaces is defined as: SPI.
[0075] Specifically, the storage module also includes two fully differential module VPX connectors P1 and P2, and each interface of the connector P1 is divided into two groups of x8 transceiver modules 0 / 1 and 2 / 3, and the interface signals of the two groups of x8 transceiver modules are defined as Aurorax8;
[0076] The interfaces of connector P2 are divided into four groups of x4 transceiver modules 0, 1, 2, and 3. Each group of x4 transceiver modules is divided into four pairs of differential signal interfaces. The interface signals of the 0th group of x4 transceiver modules are defined as: PCIE3.0x4. The four pairs of differential signal interfaces A, B, C, and D of the 2nd group of x4 transceiver modules, and the D pair of differential signal interfaces are defined as: SGMII; the four pairs of differential signal interfaces A, B, C, and D of the 3rd group of x4 transceiver modules, and the three pairs of differential signal interfaces A, B, and C are defined as: LVDS.
[0077] Specifically, the PCIE switching module also includes two fully differential module VPX connectors P1 and P2, and the interfaces of the connector P1 are divided into two groups of x8 transceiver modules 0 / 1 and 2 / 3, and the interface signals of the two groups of x8 transceiver modules are defined as PCIE3.0x8; the interfaces of the connector P2 are divided into four groups of x4 transceiver modules 0, 1, 2, 3, and the interface signals of the four groups of x4 transceiver modules are defined as: PCIE3.0x4;
[0078] The CPU module also includes two fully differential module VPX connectors P1 and P2, and the interfaces of the connector P1 are divided into two groups of x8 transceiver modules 0 / 1 and 2 / 3, which are used to interconnect the data interfaces of the switching module, and the interface signals of these two groups of x8 transceiver modules are defined as PCIE3.0x8; the interfaces of the connector P2 are divided into four groups of x4 transceiver modules 0, 1, 2, and 3, and each group of x4 transceiver modules is divided into four pairs of differential signal interfaces, and the interface signals of the second group of x4 transceiver modules are defined as: PCIE2.0x1, and the interface signals of the third group of x4 transceiver modules are defined as: SGMII;
[0079] The GPU module also includes a fully differential module VPX connector P1, which divides the interfaces of the connector P1 of the GPU module into two groups of x8 transceiver modules 0 / 1 and 2 / 3, and defines the interface signal of the 0 / 1 group of x8 transceiver modules as PCIE3.0x8.
[0080] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. According to the technical essence of the present invention, within the spirit and principles of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.
Claims
1. A method for realizing ultra-high-speed bandwidth based on a universal backplane, characterized in that: It includes a playback module, an acquisition module and a storage module. The VPX high-speed interfaces of the acquisition module and the playback module include a high-speed data interface defined by Aurorax8 signals and an extended high-speed interface defined by Aurorax4 signals. The extended high-speed interface of the acquisition module and the extended high-speed interface of the playback module are connected through the Aurorax4 signal line; The VPX high-speed interface of the storage module includes two groups of high-speed data interfaces and extended data interfaces, both of which are defined by Aurorax8 signals. The high-speed data interface of the storage module is connected to the high-speed data interface of the acquisition module through the Aurorax8 signal line, and the extended data interface of the storage module is connected to the high-speed data interface of the playback module through the Aurorax8 signal line.
2. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 1, characterized in that: The universal backplane also includes a PCIE switching module. The VPX high-speed interface of the PCIE switching module includes a high-speed serial interface defined by PCIE3.0 signals. The VPX high-speed interfaces of the playback module, the acquisition module and the storage module also include high-speed serial interfaces defined by PCIE3.0 signals. The high-speed serial interfaces of the playback module, the acquisition module and the storage module are respectively connected to the high-speed serial interface of the PCIE switching module through PCIE signal lines, so that data transmission and communication connections are carried out between the modules through the PCIE switching module.
3. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 2, characterized in that: The high-speed serial interface of the PCIE switch module can be configured to different high-speed link numbers according to the signal bandwidth requirements to increase the total bandwidth.
4. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 2, characterized in that: The high-speed serial interface of the PCIE switch module is configured as a PCIE3.0x4 high-speed serial interface defined by PCIE3.0x4 signals; The high-speed serial interface of the playback module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals, the high-speed serial interface of the storage module is configured as a PCIE3.0x4 high-speed serial interface defined by PCIE3.0x4 signals, and the PCIE3.0x4 high-speed serial interface of the PCIE switching module is respectively connected to the PCIE3.0x8 high-speed serial interface of the playback module and the PCIE3.0x4 high-speed serial interface of the storage module.
5. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 2, characterized in that: The high-speed serial interface of the PCIE switching module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals; the high-speed serial interface of the acquisition module is configured as a PCIE3.0x8 high-speed serial interface defined by PCIE3.0x8 signals, and the PCIE3.0x8 high-speed serial interface of the PCIE switching module is connected to the PCIE3.0x8 high-speed serial interface of the acquisition module.
6. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 2, characterized in that: The acquisition module and the playback module are respectively separately configured with a DSP1 module and a DSP2 module for assisting in data processing. The VPX high-speed interfaces of the acquisition module and the playback module also include a data interconnection interface defined by SRIO signals; the VPX high-speed interfaces of the DSP1 module and the DSP2 module both include a data interconnection interface defined by SRIO signals, the data interconnection interface of the acquisition module is connected to the data interconnection interface of the DSP1 module via an SRIOx4 signal line, and the data interconnection interface of the playback module is connected to the data interconnection interface of the DSP2 module via an SRIOx4 signal line.
7. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 6, characterized in that: The boards inserted into the playback module and the acquisition module are interchangeable. The playback module can be used to insert a playback board or an acquisition board, and the acquisition module can be used to insert a playback board or an acquisition board.
8. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 6, characterized in that: The VPX high-speed interfaces of the DSP1 module and the DSP2 module both include high-speed serial interfaces defined by PCIE3.0 signals. The high-speed serial interfaces of the DSP1 module and the DSP2 module are respectively connected to the high-speed serial interface of the PCIE switching module via PCIE signal lines.
9. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 6, characterized in that: The universal backplane also includes a CPU module. The VPX interface of the CPU module includes a control interface defined by PCIE2.0 signals. The VPX interfaces of the playback module, acquisition module, DSP1 module and DSP2 module also include control interfaces defined by PCIE2.0 signals. The control interface of the CPU module is respectively connected to the control interfaces of the playback module, acquisition module, DSP1 module and DSP2 module through PCIE2.0 signal lines.
10. The method for realizing ultra-high-speed bandwidth based on a universal backplane according to claim 6, characterized in that: The VPX interfaces of the acquisition module, playback module, storage module, DSP1 module and DSP2 module also include LVDS differential signal interfaces. The differential signal interfaces between the acquisition module, playback module, storage module, DSP1 module and DSP2 modules are respectively connected to each other through LVDS signal lines.
