CN101764760A - Multilink message capturing method, and method and system for processing multilink message - Google Patents
Multilink message capturing method, and method and system for processing multilink message Download PDFInfo
- Publication number
- CN101764760A CN101764760A CN201010131381A CN201010131381A CN101764760A CN 101764760 A CN101764760 A CN 101764760A CN 201010131381 A CN201010131381 A CN 201010131381A CN 201010131381 A CN201010131381 A CN 201010131381A CN 101764760 A CN101764760 A CN 101764760A
- Authority
- CN
- China
- Prior art keywords
- message
- loop buffer
- module
- analyzed
- ordering
- 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.)
- Granted
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention relates to a multilink message capturing method, each link is respectively corresponding to a circular buffer in a system kernel layer, a plurality of links are corresponding to a sequencing circular queue in the system kernel layer; the method comprises the steps: writing the messages resulted from each link into the circular buffers to which each link is corresponding, and writing memory addresses of the messages in the circular buffers thereof into the sequencing circular queue according to a synchronization mechanism adopted by the sequencing circular queue. The embodiment of the invention further provides a method for processing multilink message and a system corresponding thereto. The method and the system according to the embodiment of the invention can completely avoid the participation of a central processing unit in the process of capturing network message, achieve no loss of message during the capture in network environment more than kilomega, lessen the load of multilink message capturing system and enhance the performance of capturing message.
Description
Technical field
The present invention relates to the communications field, relate in particular to a kind of multilink message capturing method, multi-link messages processing method and corresponding system thereof.
Background technology
Along with the high speed development of the Internet and network exchange hardware, the data throughout of local net network outlet increases day by day.For example, mechanisms such as at present a lot of large enterprises, school, the data throughout of the core switch that it adopts is all greater than gigabit.Because such data throughout has surpassed the limit that traditional PCI-Express is caught bag, the network audit product that adopts PCI-Express to catch bag is used and difficulty occurred.For addressing the above problem, audit industry in the Internet generally adopts zero-copy to catch network message, to reach the effect that improves the packet capturing rate, perhaps, adopt processing speed hardware faster, the Intel preferable as performance (Intel) e1000e network interface card grasps network message, perhaps, strengthens the reception packet buffer and alleviates the performance loss that the frequent interruption of express network message brings.
But, in the face of the network data throughput that increases day by day, even use zero duplication technology and Intele1000e network interface card simultaneously, also unavoidable lost data packets, in addition, increasing reception packet buffer also can only solve the problem that speed difference brought between packet capturing speed and the message analysis speed, but can not solve packet capturing speed and not catch up with the bottleneck problem of actual mirror as data speed.
Summary of the invention
Embodiment of the invention technical problem to be solved is, a kind of multilink message capturing method, multi-link messages processing method, multilink message capturing system and multi-link messages treatment system are provided, can realize zero participation of central processing unit in the network message acquisition procedure, in network environment, catch message and do not lose greater than gigabit, alleviated the load of multilink message capturing system, improved the performance of catching message, when application layer was carried out the multi-link messages analysis, the load of application layer process program alleviated greatly in addition.
For solving the problems of the technologies described above, the embodiment of the invention adopts following technical scheme:
A kind of multilink message capturing method, every link be a loop buffer in the corresponding system kernel layer respectively, an ordering circle queue in the corresponding system kernel layer of multilink, and this method comprises:
To write the loop buffer of every link correspondence from the message that every link obtains,
Synchronization mechanism according to the ordering circle queue adopts writes the ordering circle queue with message at the memory address of its loop buffer.
A kind of multi-link messages processing method, every link be a loop buffer in the corresponding system kernel layer respectively, an ordering circle queue in the corresponding system kernel layer of multilink, and this method comprises:
To write the loop buffer of every link correspondence from the message to be analyzed that every link obtains, and foundation is used for each message to be analyzed is mapped to first mapped file that application space forms first buffer culture;
Synchronization mechanism according to the ordering circle queue adopts writes the ordering circle queue with message to be analyzed at the memory address of its loop buffer, and foundation is used for each memory address is mapped to second mapped file that application space forms second buffer culture;
According to first mapped file and second mapped file, each message to be analyzed and memory address thereof are mapped to application space, obtain first buffer culture and second buffer culture;
Start and analyze thread reads each message to be analyzed successively from second buffer culture memory address,
According to the memory address of each message to be analyzed, from first buffer culture, read successively and analyze each message to be analyzed, obtain analysis result information.
A kind of multilink message capturing system, comprise a plurality of network interface cards drive catch the bag module, a plurality of in the system kernel layer, be provided with loop buffer catch bag loop buffer administration module and an ordering loop buffer administration module that in the system kernel layer, is provided with the ordering circle queue, the corresponding respectively network interface card of every link drives catches bag module, one and catches bag loop buffer administration module, the corresponding ordering loop buffer administration module of multilink, wherein:
Network interface card drives and catches the bag module, is used for obtaining message from its respective links;
Catch bag loop buffer administration module, be used for its corresponding network interface card driven and catch the message that the bag module sends to and write corresponding loop buffer;
Ordering loop buffer administration module is used for the synchronization mechanism according to the employing of ordering circle queue, each network interface card is driven the message of catching the bag module records write the ordering circle queue at the memory address of its loop buffer.
A kind of multi-link messages treatment system, comprise a plurality of network interface cards drive catch the bag module, a plurality of in the system kernel layer, be provided with loop buffer catch bag loop buffer administration module, an ordering loop buffer administration module and an application layer that is provided with the ordering circle queue in the system kernel layer is read packet analysis module, the corresponding respectively network interface card of every link drives catches bag module, one and catches bag loop buffer administration module, the corresponding ordering loop buffer administration module of multilink, wherein:
Network interface card drives and catches the bag module, is used for obtaining message to be analyzed from its respective links;
Catch bag loop buffer administration module, be used for its corresponding network interface card driven and catch the message to be analyzed that the bag module sends to and write corresponding loop buffer, and set up and be used for message to be analyzed is mapped to first mapped file that application space forms first buffer culture;
Ordering loop buffer administration module, be used for synchronization mechanism according to the employing of ordering circle queue, each network interface card is driven the message to be analyzed of catching the bag module records write the ordering circle queue, and foundation is used for memory address is mapped to second mapped file that application space forms second buffer culture at the memory address of its loop buffer;
Application layer is read packet analysis module, be used for be mapped to each message to be analyzed and memory address thereof of application space respectively according to first mapped file, second mapped file, start and analyze thread reads each message to be analyzed successively from second buffer culture memory address, and according to the memory address of each message to be analyzed, from first buffer culture, read successively and analyze each message to be analyzed, obtain analysis result information.
The beneficial effect of the embodiment of the invention is:
By a kind of multilink message capturing method and corresponding system are provided, every link is a loop buffer in the corresponding system kernel layer respectively, an ordering circle queue in the corresponding system kernel layer of multilink, this method will write the loop buffer of every link correspondence from the message that every link obtains, and, message is write the ordering circle queue at the memory address of its loop buffer according to the synchronization mechanism that the ordering circle queue adopts.Like this, can realize that zero of central processing unit in the network message acquisition procedure participates in, in network environment, catch message and do not lose, alleviate the load of multilink message capturing system, improve the performance of catching message greater than gigabit; In addition, in multi-link messages processing method that adopts the embodiment of the invention and corresponding system, when application layer was carried out the multi-link messages analysis, the load of application layer process program alleviated greatly.
Below in conjunction with accompanying drawing the embodiment of the invention is described in further detail.
Description of drawings
Fig. 1 is the main flow chart of the multi-link messages processing method of the embodiment of the invention;
Fig. 2 is the primary structure figure of the multi-link messages treatment system of the embodiment of the invention.
Embodiment
The embodiment of the invention provides a kind of multilink message capturing method and multilink message capturing system, a loop buffer in the corresponding system kernel layer of its every of relating to link difference, an ordering circle queue in the corresponding system kernel layer of multilink, it mainly is the loop buffer that the message that will obtain from every link writes every link correspondence, and according to the synchronization mechanism that adopts between the ordering circle queue, message is write the ordering circle queue at the memory address of its loop buffer, like this, can realize zero participation of central processing unit in the network message acquisition procedure, in network environment, catch message and do not lose greater than gigabit, alleviate the load of multilink message capturing system, improved the performance of catching message; In addition, the embodiment of the invention also provides a kind of multi-link messages processing method and multi-link messages treatment system, it is mainly on above-mentioned multilink message capturing method basis, the message memory address to be analyzed and the message to be analyzed of being come by mapping in the application program application programs space read successively, analyze reading the message to be analyzed that obtains at last, obtain analysis result information, like this, when application layer is carried out the multi-link messages analysis, the load of application layer process program alleviates greatly, more effectively adapt to various real network environment, improved the analysis efficiency of multi-link messages greatly.
Multi-link messages processing method and system to the embodiment of the invention describes below by a specific embodiment, and the multilink message capturing method and system to the embodiment of the invention describe simultaneously.
Fig. 1 is the main flow chart of the multi-link messages processing method of the embodiment of the invention, this method is based on many auditing the networks that are used for the link composition of message transmissions, and multi-link messages treatment system as shown in Figure 2, this system comprises by a plurality of network interface cards drivings catches bag module 201, a plurality of in the system kernel layer, be provided with loop buffer catch bag loop buffer administration module 202, a multilink message capturing system that in the system kernel layer, is provided with ordering loop buffer administration module 203 compositions of ordering circle queue, and an application layer is read packet analysis module 204, wherein, every the corresponding respectively network interface card of link drives and catches bag module 201, catch bag loop buffer administration module 202 for one, the corresponding ordering loop buffer administration module 203 of multilink, application layer read packet analysis module 204 corresponding respectively each catch bag loop buffer administration module 202 and an ordering loop buffer administration module 203, with reference to Fig. 1, this method mainly comprises:
101, each network interface card drives catches bag module 201 from being obtained message to be analyzed by the respective links of auditing the network mirror image gained, particularly, bag module 201 is caught in each network interface card driving can be according to network interface card model and the corresponding version that drives, PCI-Express driving is preferably made amendment to current trafficability performance, network interface card is caught the direct memory visit (DirectMemoryAccess that message interrupts, DMA) address that its correspondence is caught a certain memory node in the loop buffer of bag loop buffer administration module 202 is revised as in the address, and cancel it submits message to be analyzed to the upper-layer protocol stack operation, by this modification, network interface card drives when catching bag module 201 each interruptions, will catch the memory node that bag loop buffer administration module 202 is applied in the loop buffers to its correspondence, DMA address as next message to be analyzed, thereby be that next message to be analyzed is reserved memory node, simultaneously the message to be analyzed of current acquisition being adopted dma mode to be sent to its correspondence catches the memory node of having reserved in the loop buffer of bag loop buffer administration module 202 and stores, addressable each network interface card drove the memory node of catching in the bag module 201 corresponding loop buffers when application layer was read packet analysis module 204 and run application, and just can obtain message to be analyzed;
Each network interface card drives catches bag module 202 after the message Interrupt Process, the memory address information of message to be analyzed in corresponding loop buffer, the corresponding network interface card serial number information of link is submitted to corresponding ordering loop buffer administration module 203, ordering loop buffer administration module 203 writes ordering circle queue end successively to the memory address of each message to be analyzed, like this, can be when application layer is read packet analysis module 204 and run application by the ordering circle queue of visit ordering loop buffer administration module 203, obtain the position of message to be analyzed in the loop buffer of catching bag loop buffer administration module 202 of network mirror image, thereby visit message to be analyzed;
According to mentioned above, each network interface card drives catches bag module 201 and can comprise as shown in Figure 2 structure:
102, each is caught bag loop buffer administration module 202 and its corresponding network interface card is driven catches the message to be analyzed that bag module 201 sends to and write corresponding loop buffer, and foundation is used for message to be analyzed is mapped to first mapped file that application space forms first buffer culture, particularly, each catches bag loop buffer administration module 202 according to the parameter of importing into, catch message that bag module 201 sends and distribute memory node the loop buffer for driving from network interface card, and set up first mapped file, this first mapped file can be the proc Memory Mapping File and its, catch bag loop buffer administration module 202 initialized the time, the loop buffer of application fixed size in the system kernel layer, each network interface card drives catches bag module 201 and catches message to be analyzed and produce when interrupting, will catch space of bag loop buffer administration module 202 applications to its correspondence, it is memory node, as the DMA address of next message to be analyzed, catch the vacant memory node of bag loop buffer administration module 202 traversal searching in loop buffer and catch bag module 201 to offer corresponding network interface card driving;
On the other hand, each catches the mapping function of bag loop buffer administration module 202 by the proc file system, according to above-mentioned first mapped file, each message to be analyzed is mapped to application space, offer application access in the first buffer culture mode, this first buffer culture comprises each content of message to be analyzed;
103, the synchronization mechanism that ordering loop buffer administration module 203 adopts according to the ordering circle queue, each message to be analyzed is write the ordering circle queue at the memory address of its loop buffer, and foundation is used for memory address is mapped to second mapped file that application space forms second buffer culture, particularly, in order to sort according to the time to the message on the multilink, thereby offering application layer reads packet analysis module 204 and handles, this method has adopted an ordering loop buffer administration module, catch bag module 201 and catch message to be analyzed and produce when interrupting when each network interface card drives, the Interrupt Process function is delivered to ordering loop buffer administration module 203 to the memory address of the current message to be analyzed of storage, the write operation of the ordering circle queue of ordering loop buffer administration module 203 can adopt spin lock synchronization mechanism or other synchronization mechanisms to carry out synchronously, each catches bag loop buffer administration module 202 according to the sequencing that message to be analyzed writes, and message to be analyzed is placed into the memory node of the loop buffer of last time having applied for;
On the other hand, ordering loop buffer administration module 203 is by the mapping function of proc file system, according to above-mentioned second mapped file, the memory address of each message to be analyzed is mapped to application space, offer application access in the second buffer culture mode, the memory address that comprises each message to be analyzed in second buffer culture, like this, application layer is read packet analysis module 204 by visiting the memory address that writes down in this second buffer culture, sequencing according to message arrival to be analyzed, obtain the position of the message to be analyzed of each link, thereby obtain content of message to be analyzed;
Each network interface card drive catch bag module 201, each catch bag loop buffer administration module 202, ordering loop buffer administration module 203 according to above-mentioned 101,102,103 step process after, application layer is read packet analysis module 204 and can be carried out centralized Analysis to the message to be analyzed of multilink correspondence and handle;
104, application layer is read 204 pairs of packet analysis modules respectively according to first mapped file, second mapped file is mapped to each message to be analyzed and memory address thereof of application space, start and analyze thread reads each message to be analyzed successively from second buffer culture memory address, and according to the memory address of each message to be analyzed, from first buffer culture, read and analyze each message to be analyzed successively, obtain analysis result information, particularly, after application layer is read packet analysis module 204 and is carried out initialization according to configuration file, read first buffer culture and second buffer culture on a plurality of links, and they are mapped to application space application program can be visited;
Application layer is read packet analysis module 204 and is started a plurality of analysis threads, from second buffer culture, read the memory address of message to be analyzed successively, read the memory node of the message to be analyzed in first buffer culture again according to memory address, and each memory node analyzed, put flag bit, finish analytic process.
Need to prove that the multilink message capturing method of the embodiment of the invention and multi-link messages processing method can perhaps, be divided two links of single-link gained based on the multilink of dividing according to VLAN based on employing one picture.
In addition, one of ordinary skill in the art will appreciate that all or part of flow process that realizes in the foregoing description method, be to instruct relevant hardware to finish by program, described program can be stored in the computer-readable recording medium, this program can comprise the flow process as the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-OnlyMemory, ROM) or at random store memory body (RandomAccessMemory, RAM) etc.
The above is the specific embodiment of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also are considered as protection scope of the present invention.
Claims (10)
1. a multilink message capturing method is characterized in that, every link is a loop buffer in the corresponding system kernel layer respectively, an ordering circle queue in the corresponding system kernel layer of multilink, and this method comprises:
To write the loop buffer of every link correspondence from the message that every link obtains,
Synchronization mechanism according to the ordering circle queue adopts writes the ordering circle queue with message at the memory address of its loop buffer.
2. the method for claim 1 is characterized in that, the loop buffer that will write every link correspondence from the message that every link obtains is specially:
Adopt the direct memory access mode will write the memory node of reserving the corresponding loop buffer of every link from the message that every link obtains.
3. the method for claim 1 is characterized in that, the ordering circle queue adopts the spin lock synchronization mechanism.
4. as each described method in the claim 1 to 3, it is characterized in that this method is based on the multilink of dividing according to VLAN, perhaps, this method is based on adopting the one picture to divide two links of single-link gained.
5. a multi-link messages processing method is characterized in that, every link is a loop buffer in the corresponding system kernel layer respectively, an ordering circle queue in the corresponding system kernel layer of multilink, and this method comprises:
To write the loop buffer of every link correspondence from the message to be analyzed that every link obtains, and foundation is used for each message to be analyzed is mapped to first mapped file that application space forms first buffer culture;
Synchronization mechanism according to the ordering circle queue adopts writes the ordering circle queue with message to be analyzed at the memory address of its loop buffer, and foundation is used for each memory address is mapped to second mapped file that application space forms second buffer culture;
According to first mapped file and second mapped file, each message to be analyzed and memory address thereof are mapped to application space, obtain first buffer culture and second buffer culture;
Start and analyze thread reads each message to be analyzed successively from second buffer culture memory address,
According to the memory address of each message to be analyzed, from first buffer culture, read successively and analyze each message to be analyzed, obtain analysis result information.
6. method as claimed in claim 5 is characterized in that, this method is based on the multilink of dividing according to VLAN, and perhaps, this method is based on adopting the one picture to divide two links of single-link gained.
7. multilink message capturing system, it is characterized in that, comprise a plurality of network interface cards drive catch the bag module, a plurality of in the system kernel layer, be provided with loop buffer catch bag loop buffer administration module and an ordering loop buffer administration module that in the system kernel layer, is provided with the ordering circle queue, the corresponding respectively network interface card of every link drives catches bag module, one and catches bag loop buffer administration module, the corresponding ordering loop buffer administration module of multilink, wherein:
Network interface card drives and catches the bag module, is used for obtaining message from its respective links;
Catch bag loop buffer administration module, be used for its corresponding network interface card driven and catch the message that the bag module sends to and write corresponding loop buffer;
Ordering loop buffer administration module is used for the synchronization mechanism according to the employing of ordering circle queue, each network interface card is driven the message of catching the bag module records write the ordering circle queue at the memory address of its loop buffer.
8. system as claimed in claim 7 is characterized in that, the network interface card driving is caught the bag module and comprised:
Receiver module is used for obtaining message from its respective links;
First delivery module, the message that is used for that receiver module is obtained adopt the direct memory access mode to be sent to its correspondence and catch the memory node that the loop buffer of bag loop buffer administration module reserves and store;
Second delivery module, the message that is used for writing down is sent to the ordering circle queue of ordering loop buffer queue management module at the memory address of its loop buffer.
9. system as claimed in claim 7 is characterized in that, described synchronization mechanism is the spin lock synchronization mechanism.
10. multi-link messages treatment system, it is characterized in that, comprise a plurality of network interface cards drive catch the bag module, a plurality of in the system kernel layer, be provided with loop buffer catch bag loop buffer administration module, an ordering loop buffer administration module and an application layer that is provided with the ordering circle queue in the system kernel layer is read packet analysis module, the corresponding respectively network interface card of every link drives catches bag module, one and catches bag loop buffer administration module, the corresponding ordering loop buffer administration module of multilink, wherein:
Network interface card drives and catches the bag module, is used for obtaining message to be analyzed from its respective links;
Catch bag loop buffer administration module, be used for its corresponding network interface card driven and catch the message to be analyzed that the bag module sends to and write corresponding loop buffer, and set up and be used for message to be analyzed is mapped to first mapped file that application space forms first buffer culture;
Ordering loop buffer administration module, be used for synchronization mechanism according to the employing of ordering circle queue, each network interface card is driven the message to be analyzed of catching the bag module records write the ordering circle queue, and foundation is used for memory address is mapped to second mapped file that application space forms second buffer culture at the memory address of its loop buffer;
Application layer is read packet analysis module, be used for be mapped to each message to be analyzed and memory address thereof of application space respectively according to first mapped file, second mapped file, start and analyze thread reads each message to be analyzed successively from second buffer culture memory address, and according to the memory address of each message to be analyzed, from first buffer culture, read successively and analyze each message to be analyzed, obtain analysis result information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101313819A CN101764760B (en) | 2010-03-24 | 2010-03-24 | Multilink message capturing method, and method and system for processing multilink message |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101313819A CN101764760B (en) | 2010-03-24 | 2010-03-24 | Multilink message capturing method, and method and system for processing multilink message |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101764760A true CN101764760A (en) | 2010-06-30 |
| CN101764760B CN101764760B (en) | 2011-11-16 |
Family
ID=42495752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101313819A Expired - Fee Related CN101764760B (en) | 2010-03-24 | 2010-03-24 | Multilink message capturing method, and method and system for processing multilink message |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101764760B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103856474A (en) * | 2012-11-29 | 2014-06-11 | 北京千橡网景科技发展有限公司 | Method and device for processing network data |
| CN105337976A (en) * | 2015-11-06 | 2016-02-17 | 西安交大捷普网络科技有限公司 | Real-time high-efficiency database audit realization method |
| CN105577567A (en) * | 2016-01-29 | 2016-05-11 | 国家电网公司 | Network data packet parallel processing method based on Intel DPDK |
| WO2022100310A1 (en) * | 2020-11-12 | 2022-05-19 | 苏州浪潮智能科技有限公司 | Method and apparatus for managing network card data packet buffering, and terminal and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007038856A1 (en) * | 2005-10-05 | 2007-04-12 | Nortel Networks Limited | Provider link state bridging |
| CN101043510A (en) * | 2007-04-29 | 2007-09-26 | 中兴通讯股份有限公司 | PCI-E bus distributed system intra-board and inter-board data communication method |
| CN101150485A (en) * | 2007-11-15 | 2008-03-26 | 曙光信息产业(北京)有限公司 | A management method for network data transmission of zero copy buffer queue |
-
2010
- 2010-03-24 CN CN2010101313819A patent/CN101764760B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007038856A1 (en) * | 2005-10-05 | 2007-04-12 | Nortel Networks Limited | Provider link state bridging |
| CN101043510A (en) * | 2007-04-29 | 2007-09-26 | 中兴通讯股份有限公司 | PCI-E bus distributed system intra-board and inter-board data communication method |
| CN101150485A (en) * | 2007-11-15 | 2008-03-26 | 曙光信息产业(北京)有限公司 | A management method for network data transmission of zero copy buffer queue |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103856474A (en) * | 2012-11-29 | 2014-06-11 | 北京千橡网景科技发展有限公司 | Method and device for processing network data |
| CN105337976A (en) * | 2015-11-06 | 2016-02-17 | 西安交大捷普网络科技有限公司 | Real-time high-efficiency database audit realization method |
| CN105577567A (en) * | 2016-01-29 | 2016-05-11 | 国家电网公司 | Network data packet parallel processing method based on Intel DPDK |
| CN105577567B (en) * | 2016-01-29 | 2018-11-02 | 国家电网公司 | Network packet method for parallel processing based on Intel DPDK |
| WO2022100310A1 (en) * | 2020-11-12 | 2022-05-19 | 苏州浪潮智能科技有限公司 | Method and apparatus for managing network card data packet buffering, and terminal and storage medium |
| US11777873B1 (en) | 2020-11-12 | 2023-10-03 | Inspur Suzhou Intelligent Technology Co., Ltd. | Method and apparatus for managing buffering of data packet of network card, terminal and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101764760B (en) | 2011-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101841470B (en) | High-speed capturing method of bottom-layer data packet based on Linux | |
| DE112020002491T5 (en) | SYSTEM AND METHODS TO FACILITATE DYNAMIC COMMAND MANAGEMENT IN A NETWORK INTERFACE CONTROLLER (NIC) | |
| CN105577567B (en) | Network packet method for parallel processing based on Intel DPDK | |
| CN104521198B (en) | For the system and method for Virtual Ethernet binding | |
| CN110022267A (en) | Processing method of network data packets and device | |
| CN103412786B (en) | High performance server architecture system and data processing method thereof | |
| CN101764760B (en) | Multilink message capturing method, and method and system for processing multilink message | |
| CN102123090B (en) | IP (Internet protocol) fragment processing method based on two-level table storage and transport layer information inquiry | |
| US9529622B1 (en) | Systems and methods for automatic generation of task-splitting code | |
| CN101267361A (en) | A high-speed network data packet capturing method based on zero duplication technology | |
| US20210211467A1 (en) | Offload of decryption operations | |
| US20040078525A1 (en) | Free memory manager scheme and cache | |
| US20050021558A1 (en) | Network protocol off-load engine memory management | |
| US9690713B1 (en) | Systems and methods for effectively interacting with a flash memory | |
| CN104133777B (en) | A kind of shared memory systems and its application method | |
| US11677676B1 (en) | Shared traffic manager | |
| WO2011015055A1 (en) | Method and system for storage management | |
| US11847108B2 (en) | System and method for capturing data to provide to a data analyser | |
| WO2016019554A1 (en) | Queue management method and apparatus | |
| CN107979490A (en) | The recording method of daily record data and server cluster | |
| CN109117386A (en) | A kind of system and method for network remote read-write secondary storage | |
| CN107241305A (en) | A kind of network protocol analysis system and its analysis method based on polycaryon processor | |
| US9594688B1 (en) | Systems and methods for executing actions using cached data | |
| US8630286B2 (en) | System and method for improving multicast performance in banked shared memory architectures | |
| CN107066340A (en) | A kind of fast packet receiving of the no trunk based on LINUX, method and apparatus of giving out a contract for a project |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20111116 |