CN111475459A - 9U VPX data storage server and data processing method - Google Patents
9U VPX data storage server and data processing method Download PDFInfo
- Publication number
- CN111475459A CN111475459A CN202010258621.5A CN202010258621A CN111475459A CN 111475459 A CN111475459 A CN 111475459A CN 202010258621 A CN202010258621 A CN 202010258621A CN 111475459 A CN111475459 A CN 111475459A
- Authority
- CN
- China
- Prior art keywords
- blade
- data storage
- data
- storage server
- optical fiber
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7803—System on board, i.e. computer system on one or more PCB, e.g. motherboards, daughterboards or blades
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
- G06F13/4221—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being an input/output bus, e.g. ISA bus, EISA bus, PCI bus, SCSI bus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Human Computer Interaction (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a 9U VPX data storage server and a data processing method. The data storage server includes: the system comprises a 9U VPX case, a VPX back plate with a plurality of slot positions and a domestic Feiteng computing blade, a data storage blade, a multi-port optical fiber interface blade and a RapidIO exchange blade, wherein the VPX back plate is positioned in the case and is provided with a plurality of slot positions; the multi-port optical fiber interface blade, the domestic Feiteng computing blade, the RapidIO switching blade and the data storage blade are interconnected through a double-star RapidIO switching network. The data storage server provided by the invention is an autonomous, controllable, high-performance, high-speed and real-time recording and playback data storage server, and can be widely applied to various radar recorders and optical fiber recording equipment.
Description
Technical Field
The invention relates to the technical field of storage equipment, in particular to a 9U VPX data storage server and a data processing method.
Background
With the improvement of the detection precision of equipment such as satellite navigation and radar, more and more data need to be stored and recorded, and the performance requirement on a data storage server is higher and higher.
For example, chinese patent document CN105759247A discloses a radar integrated recorder, wherein the radar integrated recorder uses a PCI bus, the bandwidth is 133MBps, and the speed is slow, or uses F L ASH as a storage unit, while most storage systems using F L ASH use an asynchronous operation mode, and the storage system has small storage capacity, low storage speed, and small data bandwidth, and cannot satisfy the functions of processing, distributing, storing, and playing back high-speed real-time data information.
Disclosure of Invention
In order to solve the above problems in the prior art, the present invention provides a 9U VPX data storage server and a data processing method.
The invention provides a 9U VPX data storage server, comprising: the system comprises a 9U VPX case, a VPX back plate with a plurality of slot positions and a domestic Feiteng computing blade, a data storage blade, a multi-port optical fiber interface blade and a RapidIO exchange blade, wherein the VPX back plate is positioned in the case and is provided with a plurality of slot positions;
the multi-port optical fiber interface blade, the domestic Feiteng computing blade, the RapidIO switching blade and the data storage blade are interconnected through a double-star RapidIO switching network.
Further, the domestic Feiteng computing blade adopts an FT-1500A/16 core processor.
Further, the processors of the data storage blades employ xc7z045 chips.
Further, the P L end of the xc7z045 chip outputs two SRIO X4 buses, and the SRIO X4 buses are interconnected with the RaipdIO switching blade, and each lane bandwidth of the SRIO interface on the SRIO X4 bus is 5 Gbps.
Further, the P L end of the xc7z045 chip also outputs 8 SATA3.0 interfaces for connecting msata disks.
Further, the RapidIO switch blade, the multi-port optical fiber interface blade, the domestic Feiteng computing blade and the data storage blade are respectively interconnected through at least 8-path 5Gbps SRIOX4 buses.
Further, the controller of the multi-port fiber optic interface blade employs an xc7v690t chip.
The invention provides a data processing method based on the data storage server, which comprises the following steps: a step of data storage and/or reading, wherein:
the data record storage step specifically comprises the following steps:
the data to be stored enters the multi-port optical fiber interface blade through the optical fiber interface;
the multiport optical fiber interface blade forwards the data to be stored to a RapidIO switching blade;
the RapidIO switching blade forwards the data to be stored to a data storage blade;
the data playback reading step specifically comprises:
reading data to be read from a data storage blade, and forwarding the data to be read to a RapidIO switching blade;
the RapidIO switching blade forwards the data to be read to a multi-port optical fiber interface blade;
and reading the data to be read from the multi-port fiber optic interface blade through a fiber optic interface.
Further, still include: and the upper computer selects a designated file from the data storage blade, forwards the designated file to the domestic Feiteng computing blade for processing, and stores the file processed by the domestic Feiteng computing blade to the data storage blade.
Further, still include: and the upper computer selects the appointed file from the data storage blade to upload and/or download.
The invention has the beneficial effects that:
(1) the 9U VPX data storage server provided by the invention adopts two SRIO X4 buses, and the bandwidth of each lane is 5Gbps, so that the data storage server can provide the bus bandwidth of 5GBps (40 Gbps);
(2) by adopting the VPX case, the slot definition conforms to Vita65-2010(OpenVPX specification), the slot definition has universality, interchangeability and expansibility, and the slots of the functional blades can be mixed and inserted, so that the requirement of adjusting the main control board and the storage board according to the requirement can be met.
(3) Compared with the existing storage system adopting the x86 architecture mainboard, the invention adopts the domestic Feiteng FT1500A/16 processor as the main controller, and can improve the autonomous safe controllable storage capacity.
(4) The data storage blade in the invention can be inserted with 8 msata disks, the local disk storage speed can reach 3.2GBps, and the maximum storage capacity can reach 8 TB.
(5) The data storage server provided by the invention can be inserted into a plurality of data storage blades according to the user requirements, and the data bandwidth and the storage capacity are multiplied.
(6) In the invention, cluster software can be erected on the kylin operating system of the domestic Feiteng computing blade, and when 4 domestic Feiteng computing blades are adopted, the theoretical peak floating point computing capacity can reach 384GF L OPS.
(7) Each optical fiber interface blade in the invention can expand 16 paths of optical fiber interfaces, and 2 optical fiber interface blades can expand up to 32 paths of optical fiber interfaces for access and can expand as required.
(8) In the invention, a redundant double star RapidIO switching network is adopted, once the main switching network fails, the service can be smoothly transferred to a standby switching network.
(9) The whole data storage server provided by the invention has the recording and playback bandwidth as high as 2.5GBps, and can meet the recording, storage and playback requirements of mass real-time data.
Drawings
Fig. 1 is a schematic structural diagram of a 9U VPX data storage server according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a data processing method based on a 9U VPX data storage server according to an embodiment of the present invention;
fig. 3 is a block diagram of a software architecture of a 9U VPX-based data storage server according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a 9U VPX data storage server, which comprises: the system comprises a 9U VPX case, a VPX back plate with a plurality of slot positions and a domestic Feiteng computing blade, a data storage blade, a multi-port optical fiber interface blade and a RapidIO exchange blade, wherein the VPX back plate is positioned in the case and is provided with a plurality of slot positions; the multi-port optical fiber interface blade, the domestic Feiteng computing blade, the RapidIO switching blade and the data storage blade are interconnected through a double-star RapidIO switching network.
As an implementation manner, in this embodiment, a 19-inch overhead 9U VPX chassis (width 482.6mm, height 400mm, and depth 330mm) is used, and as shown in fig. 1, the VPX backplane includes 8 functional slots and 2 swap slots. The 8 functional slots are respectively as follows: 4 slot positions for inserting a domestic Feiteng computing blade, 2 slot positions for inserting a data storage blade and 2 slot positions for inserting a multi-port optical fiber interface blade; the 2 exchange slots are used for inserting RapidIO exchange blades. Correspondingly, the data storage server in this embodiment includes 4 domestic feiteng computing blades, 2 data storage blades, 2 multiport fiber interface blades, and 2 RapidIO switch blades.
In addition, the data storage server in this embodiment provides necessary debugging interfaces such as a serial port and a network port to the outside for communicating with an upper computer.
On the basis of the embodiment, the FT-1500A/16 core processor is adopted by the domestic Feiteng computing blade, and the theoretical peak value capacity computing performance of the processor is 96GFlops @1.5GHz, so that the computing capacity of the complete machine is not less than 380Gflops can be sufficiently ensured by adopting 4 FT-1500A/16 core processors.
The processor of the data storage blade adopts an xc7z045 chip of xilinx company, the P L end of the xc7z045 chip outputs two SRIO X4 buses, the SRIO X4 bus is connected with the RaipdIO exchange blade, the bandwidth of each lane of an SRIO interface on the SRIO X4 bus is 5Gbps, so that the bandwidth of the two SRIO X4 interfaces can reach 40Gbps, the P L end of the xc7z045 chip also outputs 8 SATA3.0 interfaces for connecting msata disks, the capacity of each msata disk can reach 1TB, the storage bandwidth is 450MB/s, and 8 msata groups RAID0 are output, so that the storage capacity of the whole machine is not less than 8TB, the storage bandwidth is not less than 20Gbps, and the storage capacity and the bandwidth of the whole machine can be met.
The RapidIO switch blade, the multi-port optical fiber interface blade, the domestic Feiteng computing blade and the data storage blade are respectively interconnected through at least 8 paths of 5Gbps SRIO X4 buses, so that the bandwidth from other functional blades in the whole machine to the interconnection bus of the RapidIO switch blade is 5Gbps X4, and the bandwidth of a storage disk path is ensured to be more than or equal to 20 Gbps.
The controller for the multi-port fiber optic interface blade employs an xc7v690t chip from xilinx corporation. The chip can convert external 16 routes of Aurora data into internal two routes of communication SRIO X4 data, so that the 16 routes of Aurora fiber data can be allowed to be accessed, each route of Aurora data meets the bandwidth of at least 3.125Gbps, and the data is recorded into the data storage blade when the data bandwidth of 20Gbps is met.
The data storage server provided by the invention can be used for recording, storing and managing various types of data information such as radar front-end radio frequency data, task data and the like, and has the main functions of: the processing and distribution functions of high-speed real-time data information are supported; the recording, storing and playback functions of mass real-time data are supported; and the function of dynamic online reconfiguration of FPGA software is supported. Therefore, the data storage server provided by the invention is an autonomous, controllable, high-performance, high-speed and real-time recording and playback data storage server, and can be widely applied to various radar recorders and optical fiber recording equipment.
Based on the 9U VPX data storage server, an embodiment of the present invention further provides a data processing method, including: a step of data storage and/or reading. Wherein:
the data record storage step specifically comprises the following steps: the data to be stored enters the multi-port optical fiber interface blade through the optical fiber interface; the multiport optical fiber interface blade forwards the data to be stored to a RapidIO switching blade; the RapidIO switching blade forwards the data to be stored to a data storage blade;
the data playback reading step specifically comprises: reading data to be read from a data storage blade, and forwarding the data to be read to a RapidIO switching blade; the RapidIO switching blade forwards the data to be read to a multi-port optical fiber interface blade; and reading the data to be read from the multi-port fiber optic interface blade through a fiber optic interface.
The storage path is a path represented by ① serial number in FIG. 2, namely, from one or more external optical fiber interfaces- > RapidIO switching blade- > switching network- > large-capacity data storage blade, and the reading or playback path is a path represented by ① serial number, wherein the data carrying rate of the path represented by ① serial number can meet the communication rate requirement (input and output sum) of 20 Gbps.
On the basis of the above embodiment, the data processing method further includes: and the upper computer selects a designated file from the data storage blade, forwards the designated file to the domestic Feiteng computing blade for processing, and stores the file processed by the domestic Feiteng computing blade to the data storage blade.
Specifically, the upper computer calls a designated file on the data storage blade, and after the designated file is sent to the domestic Feiteng computing blade through the RapidIO switching blade, cluster software on the Feiteng computing blade is started to perform data processing, namely, a path is shown as ② serial number in FIG. 2.
On the basis of the foregoing embodiments, the data processing method further includes: and the upper computer selects the appointed file from the data storage blade to upload and/or download.
Specifically, the path carries a data rate that can meet the communication rate requirement (i.e., the sum of input and output) of 1Gbps, as indicated by reference numeral ③ in fig. 2.
As shown in fig. 3, an embodiment of the present invention further provides a software general architecture block diagram of a 9U VPX data storage server, where the software part mainly includes: the system comprises host computer management software, multi-port optical fiber interface blade software, Feiteng computing blade software, exchange network routing software and data storage blade software. Wherein:
the multi-port optical fiber interface blade software (based on Virtex-7 series FPGA) realizes interface data transmission of 16 optical ports and two SRIO X4, the optical port transmission is realized by using FPGA GTX, the link rate is adjustable, the highest rate of a single optical port is 2.5Gbps, the total of 16 effective broadband is 2Gbps considering the protocol loss, the total of 16 effective broadband is 32Gbps, the actual requirement of the whole machine on the data bandwidth of 20Gbps is met, the GTX link protocol selects AURORA protocol, selects frame mode, the AURORA protocol has flow control function, the data can not be lost under the condition that a receiving buffer area is full, the RapidIO data communication is realized in a mode of reading and writing through SRIO DMA (shared memory), the effective broadband of the two SRIO 4 buses can reach 32Gbps, the bandwidth requirement of the whole machine is met, in addition, a set of F L ASH programming logic is arranged in the FPGA, an upper computer downloads BIN file to a RapidIO exchange blade through Ethernet, and the FPGA L FPGA is configured through an SRIO instruction, thereby realizing the on-line FPGA programming function of the FPGA.
Feiteng computing blade software (based on a domestic Feiteng FT-1500A/16 core processor): the Feiteng computing blade uses a domestic Galois operating system to load a board-level SRIO driver, realizes data communication in an SRIO DMA read-write mode of RapidIO, receives and processes data from the data storage blade through a switched network route, and returns a processing result to the data storage blade.
The data storage blade software (based on Zynq-7 series FPGA) is characterized in that the functions of SRIO X4 and SATA3.0HOST are realized by FPGAIPCore at the P L end of the FPGA, an RAID array consisting of 8 SATA SSDs supports three RAID modes of RAID0, RAID1 and RAID3, the RAID modes can be set on line, the data recovery reconstruction of specified failure msata SSD can be realized by the RAID3 mode, the main functions of the PS end (based on PETA L INUX operating system) of the FPGA comprise file reading and writing, file information recording (in a data base mode), file retrieval, file playback, interaction with host computer management software and the like, the PS end software mainly comprises two processes, namely a recording and playback process and a data base management process, communication is carried out between the two processes through a process pipeline, the recording and playback process stores data received by the multi-port optical fiber interface blade into a data storage blade in a file form, the file information is sent to the data base management process through the process, the data storage blade writes the file information into the data base conveniently, each path is written in a data base, the optical path describes the data storage blade is written to the file storage blade, and the file storage blade are written to the file management process, and the file management blade, and the file management software are written to the file management software, and the file management process are written to be written to the file management software, and are written to be.
Upper computer management software: in order to facilitate the client to better view the file information, the upper computer management software comprises the contents of file retrieval, file operation, file playback, acquisition control, function test and the like. In addition, functions such as PCIE fiber card drive and network SOCKET programming are required to be added.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A 9U VPX data storage server, comprising: the system comprises a 9U VPX case, a VPX back plate with a plurality of slot positions and a domestic Feiteng computing blade, a data storage blade, a multi-port optical fiber interface blade and a RapidIO exchange blade, wherein the VPX back plate is positioned in the case and is provided with a plurality of slot positions;
the multi-port optical fiber interface blade, the domestic Feiteng computing blade, the RapidIO switching blade and the data storage blade are interconnected through a double-star RapidIO switching network.
2. The data storage server of claim 1, wherein the domestic Feiteng computing blade employs an FT-1500A/16 core processor.
3. The data storage server of claim 1, wherein the processors of the data storage blades employ xc7z045 chips.
4. The data storage server of claim 3, wherein the P L end of the xc7z045 chip outputs two SRIO X4 buses, and the SRIO X4 bus is interconnected with the RaipdIO switch blade, and each lane bandwidth of the SRIO interface on the SRIO X4 bus is 5 Gbps.
5. The data storage server of claim 3, wherein the P L end of the xc7z045 chip further outputs a SATA3.0 interface for 8 channels to connect to msata disks.
6. The data storage server of claim 1, wherein the RapidIO switch blade is interconnected with the multiport fiber optic interface blade, the domestic femtotem computing blade, and the data storage blade by at least 8 lanes of a 5Gbps SRIO X4 bus, respectively.
7. The data storage server of claim 1, wherein the controller of the multi-port fiber optic interface blade employs an xc7v690t chip.
8. A data processing method based on the data storage server of any one of claims 1 to 7, comprising: a step of data storage and/or reading, wherein:
the data record storage step specifically comprises the following steps:
the data to be stored enters the multi-port optical fiber interface blade through the optical fiber interface;
the multiport optical fiber interface blade forwards the data to be stored to a RapidIO switching blade;
the RapidIO switching blade forwards the data to be stored to a data storage blade;
the data playback reading step specifically comprises:
reading data to be read from a data storage blade, and forwarding the data to be read to a RapidIO switching blade;
the RapidIO switching blade forwards the data to be read to a multi-port optical fiber interface blade;
and reading the data to be read from the multi-port fiber optic interface blade through a fiber optic interface.
9. The data processing method of claim 8, further comprising: and the upper computer selects a designated file from the data storage blade, forwards the designated file to the domestic Feiteng computing blade for processing, and stores the file processed by the domestic Feiteng computing blade to the data storage blade.
10. The data processing method of claim 8, further comprising: and the upper computer selects the appointed file from the data storage blade to upload and/or download.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010258621.5A CN111475459A (en) | 2020-04-03 | 2020-04-03 | 9U VPX data storage server and data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010258621.5A CN111475459A (en) | 2020-04-03 | 2020-04-03 | 9U VPX data storage server and data processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111475459A true CN111475459A (en) | 2020-07-31 |
Family
ID=71749595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010258621.5A Pending CN111475459A (en) | 2020-04-03 | 2020-04-03 | 9U VPX data storage server and data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111475459A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113176850A (en) * | 2021-03-12 | 2021-07-27 | 湖南艾科诺维科技有限公司 | Shared storage disk based on SRIO interface and access method thereof |
| CN115955800A (en) * | 2023-02-27 | 2023-04-11 | 湖南博匠信息科技有限公司 | VPX data reinforcement recorder and VPX machine case |
-
2020
- 2020-04-03 CN CN202010258621.5A patent/CN111475459A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113176850A (en) * | 2021-03-12 | 2021-07-27 | 湖南艾科诺维科技有限公司 | Shared storage disk based on SRIO interface and access method thereof |
| CN115955800A (en) * | 2023-02-27 | 2023-04-11 | 湖南博匠信息科技有限公司 | VPX data reinforcement recorder and VPX machine case |
| CN115955800B (en) * | 2023-02-27 | 2023-05-12 | 湖南博匠信息科技有限公司 | VPX data reinforcement recorder and VPX chassis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9250687B1 (en) | High performance flexible storage system architecture | |
| CN102081561B (en) | Mirroring data between redundant storage controllers of a storage system | |
| US7519770B2 (en) | Disk array control device with an internal connection system for efficient data transfer | |
| US5499341A (en) | High performance image storage and distribution apparatus having computer bus, high speed bus, ethernet interface, FDDI interface, I/O card, distribution card, and storage units | |
| US8595313B2 (en) | Systems and method for simple scale-out storage clusters | |
| US10585609B2 (en) | Transfer of storage operations between processors | |
| EP0485507B1 (en) | Modular input/output system for supercomputers | |
| US8583992B2 (en) | SAS-based semiconductor storage device memory disk unit | |
| US9871704B2 (en) | High-availability computer system, working method and the use thereof | |
| CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
| US6578128B1 (en) | Address management for a shared memory region on a multi-processor controller board | |
| CN101739220A (en) | Method for designing multi-controller memory array | |
| CN103002046B (en) | Multi-system data copying remote direct memory access (RDMA) framework | |
| CN112562121A (en) | Storage data and rapid downloading method based on SRIO protocol | |
| CN111475459A (en) | 9U VPX data storage server and data processing method | |
| US11288122B1 (en) | High availability storage system | |
| US20060117159A1 (en) | Data storage system and data storage control device | |
| US7752340B1 (en) | Atomic command retry in a data storage system | |
| CN105511990B (en) | Device based on fusion architecture dual redundant degree storage control node framework | |
| CN211506479U (en) | 9U VPX data storage server | |
| US6643722B1 (en) | Data storage system having director boards with plural processors | |
| CN217640183U (en) | Storage server | |
| CN213934875U (en) | Mass data's quick uninstallation equipment | |
| CN117130537A (en) | storage server | |
| JP4571963B2 (en) | Disk array controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |