CN101299757A - Data sharing method and communication system as well as correlation equipment - Google Patents
Data sharing method and communication system as well as correlation equipment Download PDFInfo
- Publication number
- CN101299757A CN101299757A CNA2008100977374A CN200810097737A CN101299757A CN 101299757 A CN101299757 A CN 101299757A CN A2008100977374 A CNA2008100977374 A CN A2008100977374A CN 200810097737 A CN200810097737 A CN 200810097737A CN 101299757 A CN101299757 A CN 101299757A
- Authority
- CN
- China
- Prior art keywords
- data
- processing request
- data processing
- storage system
- pcie
- 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
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention discloses a data sharing method and a communications system and a device thereof, for realizing the data sharing of a plurality of host computers to reduce the cost, The embodiment method of the invention includes: receiving the data processing request transmitted by at least two host computers through the PCIE channel by the peripheral equipment expansion interface PCIE exchanger; transmitting the received data processing request to the memory system, indicating the memory system to obtain the corresponding data according to the data processing request, and indicating the memory system to perform the corresponding process on the obtained data, wherein the memory system and at least two host computers perform the address space isolation through the non-transparent port bridge NTP in the PCIE exchanger. The embodiment of the invention also provides a communication system and a PCIE exchanger. The embodiment of the invention realizes the data sharing between a plurality of host computers.
Description
Technical field
The present invention relates to communication field, relate in particular to a kind of data sharing method and communication system and relevant device.
Background technology
Along with the Internet (Internet) and in-house network (Intranet) broad application, the data message amount increases rapidly.Every main frame all had the memory device of oneself in the past, and information is isolated each other, formed information island.Continuous development along with business demand, demand to information grows with each passing day, especially the application in large data warehouse, large-scale design system, graphic system, online transacter, wideband multimedia system, telecommunication integrated service system, securities exchange system, the appearance of banking system is demanded a kind of computer system of storing big capacity information amount urgently and is occurred.These demands have been satisfied in the appearance of storage area network (SAN, StorageArea Network).
SAN is a kind of special storage networking, and it can link together various memory devices, forms the storage resources of unified management, and main frame visits these resources by optical fiber or Internet protocol (IP, Internet Protocol) network.
The many main frames of field of storage of the prior art are shared the general disk array that adopts and are linked among the storage area network SAN by optical-fibre channel (FC, Fiber Channel) interface, for multiple host provides shared memory space.
Be the share and access of realization memory device between many main frames, and give full play to the high-transmission bandwidth of fiber channel network, generally storage Sharing Management software can be installed on All hosts.Storage shareware server end main frame is installed, be metadata control (MDC, Metadata Controller) server, be in charge of access rights, the administration authority of shared volume, and the transmission of metadata control information and management, concrete system architecture is as shown in Figure 1.
In the prior art, need to realize that by FC many main frames are shared, then increased the cost of realizing that many main frames are shared.
Summary of the invention
The embodiment of the invention provides a kind of data sharing method and communication system and relevant device, realizes the data sharing of many main frames, and reduces and realize cost.
The data sharing method that the embodiment of the invention provides comprises: ancillary equipment expansion interface PCIE switch receives the data processing request that at least two main frames send by the PCIE passage; The data processing request that receives is sent to storage system, indicates described storage system to obtain corresponding data, and indicate described storage system that the described data that get access to are handled accordingly according to described data processing request; Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
The communication system that the embodiment of the invention provides comprises: the PCIE switch, be used for receiving the data processing request of main frame transmission at least by the PCIE passage, and transmit described data processing request; Storage system is used for obtaining corresponding data according to the data processing request that receives according to described data processing request, and the described data that get access to are handled accordingly; Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
The PCIE switch that the embodiment of the invention provides comprises: receiving element is used for receiving the data processing request that at least two main frames send by the PCIE passage; Retransmission unit is used for transmitting described data processing request to storage system; Nontransparent port bridge NTP is used for described storage system and described at least two main frames are carried out the address space isolation.
As can be seen from the above technical solutions, the embodiment of the invention has the following advantages:
In the embodiment of the invention, ancillary equipment expansion interface (PCIE, Peripheral ComponentInterconnect Express) switch receives the data processing request that main frame sends by the PCIE passage, send this data processing request to storage system, and utilize nontransparent port (NTP, Non-TransportPort) function realizes that the address between main frame and the storage system isolates and conversion, therefore can realize the data sharing between many main frames, and need not the participation of FC, so can realize the data sharing of many main frames, and reduce and realize cost.
Description of drawings
Fig. 1 is a system architecture schematic diagram in the prior art;
Fig. 2 is a data sharing method embodiment flow chart in the embodiment of the invention;
Fig. 3 is a communication system embodiment schematic diagram in the embodiment of the invention;
Fig. 4 is a system architecture schematic diagram in the embodiment of the invention;
Fig. 5 is a PCIE switch embodiment schematic diagram in the embodiment of the invention;
Fig. 6 is a storage system embodiment schematic diagram in the embodiment of the invention.
Embodiment
The embodiment of the invention provides a kind of data sharing method and communication system and relevant device, is used to realize the data sharing of many main frames, and reduces and realize cost.
At first introduce the data sharing method embodiment in the embodiment of the invention below, specifically comprise:
Ancillary equipment expansion interface PCIE switch receives the data processing request that at least two main frames send by the PCIE passage;
The data processing request that receives is sent to storage system, indicates described storage system to obtain corresponding data, and indicate described storage system that the described data that get access to are handled accordingly according to described data processing request;
Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
Utilize ancillary equipment expansion interface (PCIE in the present embodiment, Peripheral ComponentInterconnect Express) NTP function and PCIE switch realize that the address between main frame and storage system isolates and conversion, thereby reach the data sharing purpose, owing to need not the participation of FC, so reduced cost.
In the embodiment of the invention, storage system is receiving the data processing request that the storage transmission processing unit sends (in present embodiment and the subsequent embodiment all with input and output (IO, Input-Output) request describes as the example of data processing request, be understandable that, can also be the data processing request of other types in actual applications) afterwards, at first obtain corresponding data according to described data processing request, then by direct memory visit (DMA, Direct Memory Access) mode is handled accordingly to the described data that get access to, and adopt dma mode to carry out the storage of data and read the participation that does not need CPU, so the computing expense that can save CPU, from improving entire system IO performance.
Specific embodiment to the data sharing method in the embodiment of the invention is introduced below, need to prove, for realizing the data sharing of many main frames, then have two main frames in the present embodiment at least, present embodiment describes from the IO request to one of them main frame of the angle of PCIE switch, is understandable that, in actual applications, the PCIE switch receives the IO request that two main frames send to I haven't seen you for ages, see also Fig. 2, and data sharing method embodiment comprises in the embodiment of the invention:
201, the PCIE switch receives the data processing request that main frame sends by the PCIE passage;
In the present embodiment, when host request is carried out the read-write operation of shared data, can send the reading and writing data request to the PCIE switch by the PCIE passage that links to each other with self, if the certificate of reading, then carry the sign of the data that need read in this request, if the storage data then carry the relevant information of the data that need storage in this request.
The sign that can also comprise the transmit leg main frame in this data processing request.
202, the PCIE switch sends data processing request to the storage transmission processing unit;
PCIE receives after the data processing request of main frame transmission, this request is forwarded to the storage transmission processing unit, carry out relevant treatment with indication storage transmission processing unit, concrete processing comprises that the main frame to sending this data processing request carries out authorization check, owing to can comprise the sign of transmit leg main frame in the data processing request, so the storage transmission processing unit is according to the corresponding main frame of this sign inquiry, and whether this main frame of inquiry has authority to initiate data processing request in the permissions list that presets, if authority is arranged, then execution in step 203, if there is not authority, execute exception handling process then, be specifically as follows to main frame feedback error information, be understandable that, the abnormality processing flow process in the practical application does not limit herein.
Need to prove that the storage transmission processing unit is except carrying out the function of authorization check in the present embodiment, the function of all right executive address conversion specifically will be elaborated in subsequent embodiment.
203, the storage transmission processing unit is transmitted this data processing request to storage system;
In the present embodiment, after the authority verification was passed through, the storage transmission processing unit can be forwarded to storage system with the data processing request that the PCIE switch sends.
Need to prove that in the present embodiment, above-mentioned data processing request can be sent to storage system by the mode of address transition.
In the present embodiment, the storage transmission processing unit can not carry out authorization check to data processing request equally in the step 202, promptly give tacit consent to the main frame that all main frames of initiating data processing request are authority, then in this case, the storage transmission processing unit can be directly be forwarded to storage system with the data processing request of PCIE switch transmission.
204, storage system is obtained corresponding data according to data processing request;
Storage system is obtained the data corresponding with this request after the data processing request that receives the transmission of storage transmission processing unit.
If data read request, then storage system is according to the corresponding data of inquiry in the data of self storing that are identified at of the data of carrying in the data read request;
If data storage request, then storage system receives corresponding data from the PCIE switch, need to prove, if main frame is a data storage request to the transmission of PCIE switch, then main frame also can send the data that need storage to the PCIE switch accordingly, the PCIE switch gets access to after these data, with these data forwarding to storage system.
205, storage system is handled the data that get access to accordingly by dma mode.
Storage system has been determined after the data corresponding with data processing request, if data read request then adopts dma mode from the data of self storing this partial data to be copied to from memory in the internal memory, the main frame to correspondence sends this partial data again;
If data storage request then adopts dma mode to be copied in the memory from internal memory from the data that the PCIE switch gets access to and stores.
If data read request, then its detailed process is: storage system is sent to corresponding main frame by the NTP address transition with the The data dma mode that gets access to.
After storage system is handled the described data that get access to accordingly by dma mode, also need storage system and main frame are carried out synchronously, concrete synchronizing process can for: after DMA finishes, storage system adopts interrupt mode notice host dma to finish, be understandable that, in actual applications, can also adopt other mode realize between storage system and the main frame synchronously, concrete mode does not limit herein.
In the present embodiment, the PCIE switch receives the data processing request that main frame sends by the PCIE passage, send this data processing request to storage system, and utilize the NTP function to realize that the address between main frame and the storage system isolates and conversion, therefore can realize the data sharing between many main frames, and need not the participation of FC, thus can realize the data sharing of many main frames, and reduce and realize cost;
In addition, in the present embodiment, under the dma mode, the transfer of data from RAM to other IO equipment no longer needs central processing unit (CPU, Center Processing Unit) to coordinate.DMA has replaced CPU, allows storage medium can directly copy content in other storage mediums, so just reduced the frequent intervention of CPU in data transmission procedure, and whole data transmission procedure is finished automatically by both party, has therefore improved entire system IO performance.
Introduce the communication system embodiment in the embodiment of the invention below, see also Fig. 3, the communication system embodiment in the embodiment of the invention comprises:
Storage transmission processing unit 302 is used to receive the data processing request that PCIE switch 301 sends, and according to the described data processing request authorization check that conducts interviews, if verification is passed through, then sends described data processing request to storage system 303;
Storage transmission processing unit 302 in the present embodiment also is used for the main frame that belongs to different address fields is carried out address translation, and concrete process is elaborated in subsequent embodiment.
Need to prove, in the present embodiment,, then store transmission processing unit 302 and only play the data penetration transmission effect if need not to carry out authorization check.
In the present embodiment, PCIE switch 301 receives the data processing request that main frame sends by the PCIE passage, send this data processing request by storage transmission processing unit 302 to storage system 303, and utilize the NTP function to realize that the address between main frame and the storage system isolates and conversion, therefore can realize the data sharing between many main frames, and need not the participation of FC, thus can realize the data sharing of many main frames, and reduce and realize cost;
In addition, because storage system 303 adopts dma mode that data are handled, need not the participation of CPU, thereby improved the whole IO performance of system.
For ease of understanding, describe with an instantiation below, see also Fig. 4, in the system architecture of the embodiment of the invention:
The PCIE switch is made up of 4 PCIE Switch chips.Comprising a slice master exchange chip and 3 exchange chips, this main exchange chip can provide several ports for several main frame shared data; Inferior exchange chip provides many main frames sharing functionality, and promptly each time exchange chip can corresponding main frame.
Storage system adopts synchronous DRAM (SDRAM, Synchronous DynamicRandom Access Memory).
Be the consisting of of each module in this programme:
PCIE switch: form by 1 main exchange chip and 3 exchange chips.
Storage system: comprise 16 SDRAM below some sdram controllers and the some SDRAM composition, each sdram controller.
Storage transmission process module: the chip of handling the storage transmission.
In this programme, the PCIE switch externally provides 3 PCIE standard cable interfaces, and many host accesss storage system functionality is provided, and all is the PCIE interface, supports 3 Server to share the intrasystem stored information of scheme.
In this programme, system offers the interface of 3 outside PCIE, can allow three main frames share a storage system, but owing to belong to different address fields, if directly share, then can produce address conflict, therefore in the inferior exchange chip of PCIE switch, also have a port to support nontransparent port (NTP, Non-Transparent Port) therefore function can reach the effect that the address isolates.
After isolating through NTP, inferior exchange chip can play the effect of address translation in two address fields, has so just realized that many main frames share a storage system.
Inferior exchange chip is as the transparent exchange time spent, and initial is exactly a Fan-out function, can be a plurality of PCIE equipment of system extension.As the NTP time spent, can be operated in intelligent adapter (intelligent adaptercard) pattern, also can be operated in two main frame (Dual host) patterns, this programme system is operated in Intelligent adapter card pattern, and Dual host pattern is generally used in memory device often, uses as two primary mirror pictures backups.During intelligent adapter card pattern, reset Host reset signal receiving from the main frame of its upstream data Upstream, inferior exchange chip its inner SwitchFabric that will reset, transparent Downstream port, with the state of the Virtual side of NTP, but the state of the Link Side of the NTP that can not reset.
In this programme:
Main frame is connected to the PCIE switch by PCIE, and the data processing request of main frame (IO request) sends to the PCIE switch by the PCIE passage.
The PCIE switch is transmitted to unified processing of storage transmission process module with all IO requests, to guarantee data sync.
Storage transmission process module sends order to the sdram controller in the storage system, obtains the data of order correspondence.
Main exchange chip provides plurality of ports altogether in this programme, using in this programme only provides three main frames to share storage, also can expand again in other cases, provide more main frame to share, in storage system, what this programme was described is the situation that 32 SDRAM are arranged under 2 sdram controllers, can expand in the practical application.
In the present embodiment, the PCIE switch receives the IO request that main frame sends by the PCIE passage, and all IO request is transmitted to unified processing of storage transmission process module, institute is so that the data between the data of each main frame and the storage system keep synchronously, so improved data transmission credibility;
Secondly, because the main exchange chip in the PCIE switch can provide plurality of ports, provide share service so can expand to more main frame;
Once more, because the storage system in the present embodiment adopts the mode of sdram controller and SDRAM, and can control 32 SDRAM under each sdram controller, so can increase the memory space of storage system, and sdram controller can further expand the SDRAM of its control, helps extension storage system space easily.
Introduce the PCIE switch embodiment in the embodiment of the invention below, see also Fig. 5, PCIE switch embodiment comprises in the embodiment of the invention:
Receiving element 501 is used for receiving the data processing request that main frame sends by the PCIE passage;
NTP504 is used for described storage system and described main frame are carried out the address space isolation.
PCIE switch in the present embodiment can further include:
Data transmission unit 503 is used for sending data to be stored by the PCIE passage to storage system.
In the present embodiment, after receiving element 501 receives data processing request from main frame, by retransmission unit 502 data processing request that receives is forwarded to storage system, and the address space of storage system and main frame is isolated by NTP504, therefore can realize the data sharing between many main frames, and need not the participation of FC, thus can realize the data sharing of many main frames, and reduce and realize cost.
See also Fig. 6, storage system is executed example and is comprised in the embodiment of the invention:
Request receiving element 601 is used to receive the data processing request that the storage transmission processing unit sends;
Data capture unit 602 is used for obtaining corresponding data according to described data processing request;
Data processing unit 603 is used for by dma mode the described data that get access to being handled accordingly.
Data processing unit 603 in the present embodiment can further include:
Data-reading unit 6031 is used for reading by dma mode the data of described data processing request correspondence, and described data is sent to the main frame of correspondence;
Data storage cell 6032 is used for storing the data that described data capture unit gets access to from the PCIE switch by dma mode.
One of ordinary skill in the art will appreciate that all or part of step that realizes in the foregoing description method is to instruct relevant hardware to finish by program, described program can be stored in a kind of computer-readable recording medium, this program comprises the steps: when carrying out
Ancillary equipment expansion interface PCIE switch receives the data processing request that at least two main frames send by the PCIE passage; The data processing request that receives is sent to storage system, indicates described storage system to obtain corresponding data, and indicate described storage system that the described data that get access to are handled accordingly according to described data processing request; Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
The above-mentioned storage medium of mentioning can be a read-only memory, disk or CD etc.
More than a kind of data sharing method provided by the present invention and communication system and relevant device are described in detail, for one of ordinary skill in the art, thought according to the embodiment of the invention, part in specific embodiments and applications all can change, in sum, this description should not be construed as limitation of the present invention.
Claims (10)
1, a kind of data sharing method is characterized in that, comprising:
Ancillary equipment expansion interface PCIE switch receives the data processing request that at least two main frames send by the PCIE passage;
The data processing request that receives is sent to storage system, indicates described storage system to obtain corresponding data, and indicate described storage system that the described data that get access to are handled accordingly according to described data processing request;
Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
2, method according to claim 1 is characterized in that, the described data processing request that will receive comprises before being sent to the step of storage system:
The data processing request that receives is sent to the storage transmission processing unit, indicates described storage transmission processing unit that the main frame of initiating described data processing request is carried out authorization check.
3, method according to claim 1 is characterized in that, the described step that the data that get access to are handled accordingly comprises:
If described data processing request is a data read request, then storage system sends the described data that get access to by direct memory visit dma mode to the main frame of correspondence;
If described data processing request is a data storage request, then storage system is stored the described data that get access to by dma mode.
According to each described method in the claim 1,2 or 3, it is characterized in that 4, described data processing request is sent to described storage system by the mode of address transition;
Described storage system comprises by the step of dma mode to the described data that get access to of main frame transmission of correspondence:
Storage system is carried out the address space that address after the address transition is sent to the described The data dma mode that gets access in described main frame according to described storage transmission processing unit.
According to each described method in the claim 1,2 or 3, it is characterized in that 5, the described step that the data that get access to are handled accordingly comprises afterwards:
Storage system adopts interrupt mode notice host dma to finish.
6, a kind of communication system is characterized in that, comprising:
The PCIE switch is used for receiving the data processing request of main frame transmission at least by the PCIE passage, transmits described data processing request;
Storage system is used for obtaining corresponding data according to the data processing request that receives according to described data processing request, and the described data that get access to are handled accordingly;
Described storage system and described at least two main frames carry out address space by the nontransparent port bridge NTP in the described PCIE switch isolates.
7, communication system according to claim 6 is characterized in that, described communication system also comprises:
The storage transmission processing unit is used to receive the data processing request that the PCIE switch sends, and according to the described data processing request authorization check that conducts interviews, if verification is passed through, then sends described data processing request to storage system.
8, communication system according to claim 7 is characterized in that, described storage transmission processing unit also is used for the main frame that belongs to different address fields is carried out address translation.
9, a kind of PCIE switch is characterized in that, comprising:
Receiving element is used for receiving the data processing request that at least two main frames send by the PCIE passage;
Retransmission unit is used for transmitting described data processing request to storage system;
Nontransparent port bridge NTP is used for described storage system and described at least two main frames are carried out the address space isolation.
10, PCIE switch according to claim 9 is characterized in that, described PCIE switch also comprises:
Data transmission unit is used for sending data to be stored by the PCIE passage to storage system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100977374A CN101299757A (en) | 2008-05-23 | 2008-05-23 | Data sharing method and communication system as well as correlation equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100977374A CN101299757A (en) | 2008-05-23 | 2008-05-23 | Data sharing method and communication system as well as correlation equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101299757A true CN101299757A (en) | 2008-11-05 |
Family
ID=40079437
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2008100977374A Pending CN101299757A (en) | 2008-05-23 | 2008-05-23 | Data sharing method and communication system as well as correlation equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101299757A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103117929A (en) * | 2013-01-31 | 2013-05-22 | 中国科学院计算技术研究所 | Communication method and system based on PCIe (peripheral component interconnect express) data exchange |
US8566416B2 (en) | 2011-09-23 | 2013-10-22 | Huawei Technologies Co., Ltd. | Method and system for accessing storage device |
CN103384551A (en) * | 2013-01-04 | 2013-11-06 | 华为技术有限公司 | Virtual machine communication method and system based on PCIE network, and server |
CN104111907A (en) * | 2014-06-27 | 2014-10-22 | 华为技术有限公司 | Method for accessing NVMe storage device and NVMe storage device |
WO2015010603A1 (en) * | 2013-07-22 | 2015-01-29 | Huawei Technologies Co., Ltd. | Scalable direct inter-node communication over peripheral component interconnect-express (pcie) |
WO2015043175A1 (en) * | 2013-09-29 | 2015-04-02 | 华为技术有限公司 | Server system and operation system starting method thereof, and starting management node |
CN104965677A (en) * | 2015-06-26 | 2015-10-07 | 北京百度网讯科技有限公司 | Storage system |
CN105224468A (en) * | 2014-06-09 | 2016-01-06 | 华为技术有限公司 | A kind of internal memory sharing method, isomerization hardware, main frame and heterogeneous system |
CN107656894A (en) * | 2017-09-25 | 2018-02-02 | 联想(北京)有限公司 | A kind of more host processing systems and method |
CN110099040A (en) * | 2019-03-01 | 2019-08-06 | 江苏极元信息技术有限公司 | A kind of defence method intercepting Intranet attack source based on a large amount of deployment bait host detections |
CN110532208A (en) * | 2019-07-12 | 2019-12-03 | 优刻得科技股份有限公司 | A kind of data processing method, interface conversion structure and equipment |
CN112558888A (en) * | 2021-02-25 | 2021-03-26 | 上海飞斯信息科技有限公司 | Multi-host shared storage device based on RapidIO bus link and sharing method thereof |
CN113611102A (en) * | 2021-07-30 | 2021-11-05 | 中国科学院空天信息创新研究院 | Multi-channel radar echo signal transmission method and system based on FPGA |
CN115098426A (en) * | 2022-06-22 | 2022-09-23 | 深圳云豹智能有限公司 | PCIE (peripheral component interface express) equipment management method, interface management module, PCIE system, equipment and medium |
-
2008
- 2008-05-23 CN CNA2008100977374A patent/CN101299757A/en active Pending
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8566416B2 (en) | 2011-09-23 | 2013-10-22 | Huawei Technologies Co., Ltd. | Method and system for accessing storage device |
CN103384551A (en) * | 2013-01-04 | 2013-11-06 | 华为技术有限公司 | Virtual machine communication method and system based on PCIE network, and server |
CN103384551B (en) * | 2013-01-04 | 2015-12-09 | 华为技术有限公司 | A kind of virtual machine communication method of Based PC IE network, server and system |
CN103117929B (en) * | 2013-01-31 | 2015-12-23 | 中国科学院计算技术研究所 | A kind of communication means of Based PC Ie exchanges data and system |
CN103117929A (en) * | 2013-01-31 | 2013-05-22 | 中国科学院计算技术研究所 | Communication method and system based on PCIe (peripheral component interconnect express) data exchange |
US9910816B2 (en) | 2013-07-22 | 2018-03-06 | Futurewei Technologies, Inc. | Scalable direct inter-node communication over peripheral component interconnect-express (PCIe) |
WO2015010603A1 (en) * | 2013-07-22 | 2015-01-29 | Huawei Technologies Co., Ltd. | Scalable direct inter-node communication over peripheral component interconnect-express (pcie) |
US11036669B2 (en) | 2013-07-22 | 2021-06-15 | Futurewei Technologies, Inc. | Scalable direct inter-node communication over peripheral component interconnect-express (PCIe) |
WO2015043175A1 (en) * | 2013-09-29 | 2015-04-02 | 华为技术有限公司 | Server system and operation system starting method thereof, and starting management node |
CN105224468A (en) * | 2014-06-09 | 2016-01-06 | 华为技术有限公司 | A kind of internal memory sharing method, isomerization hardware, main frame and heterogeneous system |
CN104111907A (en) * | 2014-06-27 | 2014-10-22 | 华为技术有限公司 | Method for accessing NVMe storage device and NVMe storage device |
WO2016206198A1 (en) * | 2015-06-26 | 2016-12-29 | 北京百度网讯科技有限公司 | Storage system |
CN104965677B (en) * | 2015-06-26 | 2018-04-13 | 北京百度网讯科技有限公司 | Storage system |
CN104965677A (en) * | 2015-06-26 | 2015-10-07 | 北京百度网讯科技有限公司 | Storage system |
CN107656894A (en) * | 2017-09-25 | 2018-02-02 | 联想(北京)有限公司 | A kind of more host processing systems and method |
CN110099040A (en) * | 2019-03-01 | 2019-08-06 | 江苏极元信息技术有限公司 | A kind of defence method intercepting Intranet attack source based on a large amount of deployment bait host detections |
CN110099040B (en) * | 2019-03-01 | 2021-11-30 | 江苏极元信息技术有限公司 | Defense method for detecting and intercepting intranet attack source based on mass bait deployment host |
CN110532208A (en) * | 2019-07-12 | 2019-12-03 | 优刻得科技股份有限公司 | A kind of data processing method, interface conversion structure and equipment |
CN110532208B (en) * | 2019-07-12 | 2021-05-28 | 优刻得科技股份有限公司 | Data processing method, interface conversion structure and data processing equipment |
CN112558888A (en) * | 2021-02-25 | 2021-03-26 | 上海飞斯信息科技有限公司 | Multi-host shared storage device based on RapidIO bus link and sharing method thereof |
CN113611102A (en) * | 2021-07-30 | 2021-11-05 | 中国科学院空天信息创新研究院 | Multi-channel radar echo signal transmission method and system based on FPGA |
CN115098426A (en) * | 2022-06-22 | 2022-09-23 | 深圳云豹智能有限公司 | PCIE (peripheral component interface express) equipment management method, interface management module, PCIE system, equipment and medium |
CN115098426B (en) * | 2022-06-22 | 2023-09-12 | 深圳云豹智能有限公司 | PCIE equipment management method, interface management module, PCIE system, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101299757A (en) | Data sharing method and communication system as well as correlation equipment | |
US6907457B2 (en) | Architecture for access to embedded files using a SAN intermediate device | |
US9213500B2 (en) | Data processing method and device | |
US8572342B2 (en) | Data transfer device with confirmation of write completion and method of controlling the same | |
CN102541984B (en) | File system of distributed type file system client side | |
US11741034B2 (en) | Memory device including direct memory access engine, system including the memory device, and method of operating the memory device | |
WO2018054079A1 (en) | Method for storing file, first virtual machine and namenode | |
WO2012145963A1 (en) | Data management system and method | |
JP6514329B2 (en) | Memory access method, switch, and multiprocessor system | |
US20240106893A1 (en) | Filecoin cluster data transmission method and system based on remote direct memory access | |
US20060236032A1 (en) | Data storage system having memory controller with embedded CPU | |
CN110119304A (en) | A kind of interruption processing method, device and server | |
CN104102742A (en) | High-performance mass storage system and high-performance mass storage method | |
US20230114242A1 (en) | Data processing method and apparatus and heterogeneous system | |
AU2014336967A1 (en) | Network interface | |
US20150074316A1 (en) | Reflective memory bridge for external computing nodes | |
US20240126847A1 (en) | Authentication method and apparatus, and storage system | |
WO2024051410A1 (en) | Data access method and apparatus, network interface card, readable medium, and electronic device | |
WO2023186143A1 (en) | Data processing method, host, and related device | |
CN101807170B (en) | Management method of cache of storage area network | |
WO2019223444A1 (en) | Data storage system | |
CN113296899A (en) | Transaction master machine, transaction slave machine and transaction processing method based on distributed system | |
US20080147933A1 (en) | Dual-Channel Network Storage Management Device And Method | |
US11334245B1 (en) | Native memory semantic remote memory access system | |
Bai et al. | An Analysis on Compute Express Link with Rich Protocols and Use Cases for Data Centers |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Open date: 20081105 |