CN100403739C - News transfer method based on chained list process - Google Patents
News transfer method based on chained list process Download PDFInfo
- Publication number
- CN100403739C CN100403739C CNB2006100074871A CN200610007487A CN100403739C CN 100403739 C CN100403739 C CN 100403739C CN B2006100074871 A CNB2006100074871 A CN B2006100074871A CN 200610007487 A CN200610007487 A CN 200610007487A CN 100403739 C CN100403739 C CN 100403739C
- Authority
- CN
- China
- Prior art keywords
- message
- progress
- linked list
- head pointer
- node
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Computer And Data Communications (AREA)
Abstract
The present invention discloses a method for transferring messages among processes based on a linked list, which comprises that: a main control progress creates a message queue which comprises an allocated node queue and a free node queue; a first processing progress receives messages coming from outside, free nodes are applied in the free node message queue, the head and the tail of message fields are linked into a shared memory linked list, and then, a head pointer of the linked list is sent to a second processing progress; after receiving the head pointer of the shared memory linked list transferred by the first processing progress, the second processing progress causes the pointer transferred by the first processing progress to be converted into a memory address in a process address space. The second processing progress visits the shared memory linked list in the memory address. After the processing is completed, the result is updated into the linked list, and then, the head pointer of the linked list is sent to a post-processing progress. A message transfer technology based on the pointer of the shared memory and the data structure of the linked list of the shared memory are adopted to effectively improve the transfer efficiency of messages under a stream type service processing frame.
Description
Technical field
The present invention relates to progress message transferring method, relate in particular to a kind of progress message transferring method based on chained list.
Background technology
In the prior art, service framework is made up of serviced component and message event.Wherein, serviced component can be independently process or thread, each serviced component comprises input message queue and an output message formation usually, and each serviced component knows its next service name, forms an end to end in logic service chained list thus.Message event can produce the serviced component that new message event is imported next linking behind each serviced component of flowing through.
" message " is the data unit that transmits at intercomputer.Message can be very simple, for example only comprises text-string; Also can be more complicated, may comprise embedded object.Message is sent in the formation." message queue " is the container of preserving message in the transmission course of message.Message queue manager is acted as an intermediary when message being relayed to its target from its source.The main purpose of formation provides route and guarantees the transmission of message; Message queue meeting reservation message is up to can successfully transmitting it.
Using socket SOCKET, pipeline interprocess communication (Inter-ProcessCommunication such as (FIFO), IPC) during machine-processed pass-along message, message need be transmitted between a plurality of processes, usually, each treatment progress all will be finished following processing links: read the message of reception from receive buffering area, and decoding becomes internal data structure; Finish processing procedure, preserve result; Result's coding after handling is become message flow, write the transmission buffering area and send to the opposite end process.For example shown in Figure 1, treatment progress 1 copies message data from receive buffering area, need then message data is decoded, after decoded data are handled accordingly, again message data is after treatment encoded, and copy the transmission buffering area to, so that next process (being treatment progress 2) can receive message data.Treatment progress 2, treatment progress 3 all repeat similar step and come message data is handled.
In above-mentioned traditional inter-process messages pass through mechanism, when transmitting between different processes, message all between buffering area and internal memory, be copied, and the coding of message event field, decoding, carry under the less situation of data volume in concurrent quantity of message and message event, aforesaid operations can not cause the bottleneck on the performance, but in the concurrent quantity of message in the more application of more or message event carrying data volume (as real-time charge system), aforesaid operations frequently takes place, and can cause systematic function seriously to descend.
According to foregoing description, there is following shortcoming in the technical scheme that can draw prior art: message repeatedly copies between reception buffering area, transmission buffering area and internal memory, and data transmission efficiency is low under the situation of big data quantity; Each message transmission all will be finished the process of coding or decoding, consumption systems cpu resource;
Summary of the invention
The problem that the present invention solves is in the prior art, and the processing to the message transmission between process has consumed ample resources, the situation that causes systematic function to descend.The invention provides a kind of progress message transferring method based on chained list that can improve transmission of messages efficient.
Progress message transferring method based on chained list provided by the invention comprises:
A) master control process creation shared drive district, described shared drive district is used to deposit the message data of interprocess communication;
Master control process creation message queue, described message queue comprise distribution node formation and idle node formation;
B) first treatment progress receives the message from the outside, applies for idle node from message queue, and message field is linked into sharing memory linked list from beginning to end that the head pointer that sends this chained list then is to second treatment progress;
C) after receiving the sharing memory linked list head pointer that first treatment progress transmits, second treatment progress is converted to memory address in the address space with the pointer of the first treatment progress transmission;
D) second treatment progress is visited the sharing memory linked list in the described memory address, after finishing processing the result is updated in this chained list, and the head pointer with this chained list sends to the subsequent treatment process then.
According to above-mentioned progress message transferring method, described shared drive district is the memory block of each field distribution predetermined length of message,
According to above-mentioned progress message transferring method, there are a plurality of treatment progress that the message of outside is handled.
According to above-mentioned progress message transferring method, last treatment progress in the described treatment progress is released to the information node of handling in the idle node formation and goes.
According to above-mentioned progress message transferring method, node comprises: message queue size, message queue idle node linked list head pointer and message queue be distribution node linked list head pointer.
According to above-mentioned progress message transferring method, between treatment progress, adopt first-in first-out FIFO mechanism to send the head pointer of sharing memory linked list.
According to above-mentioned progress message transferring method, the message queue node data comprises: at least one of the head pointer of sharing memory linked list, the tail pointer of sharing memory linked list and message data field number.
Especially, the present invention also provides a kind of charge system progress message transferring method based on chained list, comprising:
A) master control process creation shared drive district, described shared drive district is used to deposit the message data of interprocess communication; Master control process creation message queue, described message queue comprise distribution node formation and idle node formation;
B) authentication process receives the message from external progress, and applies for idle node from message queue, and message field is linked into sharing memory linked list from beginning to end, and the head pointer that sends this chained list then is to the charging process;
C) after receiving the sharing memory linked list head pointer that authentication process is transmitted, the charging process is converted to memory address in the address space with the pointer of authentication process transmission;
D) the charging process is visited the sharing memory linked list in the described memory address, after finishing processing the result after handling is updated in this chained list, and the head pointer that sends this chained list then is to the book keeping operation process;
E) book keeping operation process pointer is converted to the memory address in the address space, visits the sharing memory linked list in the described memory address.
Because the present invention adopts message transferring technology and sharing memory linked list data structure based on shared drive, improved the transmission efficiency of message under the stream mode service processing framework effectively.
In the present invention,, therefore alleviated the load of transfer of data between process, improved efficiency of transmission because the data of transmitting are the pointer of chained list.And, therefore can solve the performance issue that causes because of the change of message field Data Dynamic because chained list itself has insertion, deletion of node speed is very fast and flexible characteristics, and adopts the data structure of Dram chained list as message.Use list structure can make full use of the calculator memory space, realize internal memory dynamic management flexibly.
Description of drawings
Fig. 1 is that IPC mechanism message is transmitted schematic diagram in the prior art;
Fig. 2 is an example prototype system schematic diagram of the present invention;
Fig. 3 is the flow chart based on the progress message transferring method of chained list of example of the present invention;
Fig. 4 is the memory pool schematic diagram of mechanism that example of the present invention adopts;
Fig. 5 is that the charge system of the example according to the present invention is handled the framework schematic diagram;
Fig. 6 is the application schematic diagram of example of the present invention in charge system;
Fig. 7 is the flow chart of the present invention's message delivery method in charge system.
Embodiment
Fig. 2 is a prototype system schematic diagram of the present invention.As shown in the figure, described prototype system comprises master control process, treatment progress and external progress.Wherein, the master control process is responsible for creating message queue, and the process of master control is simultaneously also controlled each treatment progress.
External progress is input to message event in the treatment progress, and is safeguarded the data structure of this message queue in shared drive by master control process and treatment progress.The data structure of described message queue comprises: message queue size, message queue idle node linked list head pointer, message queue be distribution node linked list head pointer.
In the present embodiment, for convenience of description, only relate to three treatment progress: process 1, process 2 and process 3.These three treatment progress can carry out corresponding Message Processing, and are different according to different application about the detailed process of Message Processing.For example, three Message Processing processes can be " identification authentication ", " charging ", " book keeping operation " or the like the treatment progress in the communication charge system.From the message of external progress input between three treatment progress " circulation " and obtain processing.In actual applications, system can comprise more treatment progress.
Fig. 3 is the flow chart based on the progress message transferring method of sharing memory linked list of the embodiment of the invention.In step 301, shared drive district of master control process creation.During system start-up, at first by the shared drive district of an overall situation of master control process creation, the size of shared drive is calculated according to the disposal ability of system and is obtained, and also can adjust by the method for dynamic increase.
In order to improve the allocative efficiency of internal memory, native system has adopted shared drive pond mechanism to realize the distribution and the retrieval management of shared drive.As shown in Figure 4, sharing memory linked list is that the present invention dynamically changes the data structure of employing in order to solve message.Message data is made up of several fields, it in shared drive the memory block of the field distribution regular length of each message data, and these fields are linked at together from beginning to end, just formed sharing memory linked list, each distribution node deposit the head pointer that points to sharing memory linked list.For fear of the memory fragmentation that causes when distributing less memory block, the present invention has adopted memory pool technique.In memory pool, have a plurality of memory blocks, in each memory block, have the memory cell of predetermined length, as the memory block of 16 bytes, the memory block of 32 byte lengths, memory block of 4k byte length or the like.Field with message data of different length is stored in respectively in the memory block of respective length.
In step 302, master control process creation message queue.Create message queue provide can with the mutual resource of each treatment progress.Message queue is made up of a plurality of sharing memory linked list, comprising distribution node chained list and idle node chained list.
In step 303, the message that treatment progress 1 receives from the outside, and from described message queue, apply for idle node.Process 1 receives from the external progress message transmitted by IPC such as SOCKET or FIFO mechanism, and applies for an idle node from message queue.Comprise following message in the message queue node data structure: the head pointer of sharing memory linked list; The tail pointer of sharing memory linked list; The message data field number.The idle message queue node of application is to be the message application memory block after handling.
In step 304,1 pair of process is handled from the message that external progress received.Process 1 and the process of mentioning later 2, process 3 are treatment progress.Processing procedure to message is different according to different application.For example, three Message Processing processes can be " authentication ", " charging ", " book keeping operation " or the like the treatment progress in the communication charge system.
In step 305, the field of the message after process 1 will be handled is linked into sharing memory linked list from beginning to end, and the head pointer that sends this chained list then is to processing process 2.Again with reference to figure 4, field 1, field 2, field 3, field 4 are the fields after four process processes 1 are handled, wherein the length of field 1, field 3, field 4 is 32 bytes, and the length of field 2 is 64 bytes, and therefore four fields are stored in respectively in the memory block of 32 bytes and 64 byte longs.Process 1 links four fields from beginning to end, forms sharing memory linked list, adopts IPC mechanism (for example, FIFO mechanism) to send the head pointer (being the pointer of message queue node) of this chained list to process 2 then.The head pointer of this chained list is the head pointer of the sharing memory linked list of process 1 idle node of being applied for, owing to deposited the head pointer that points to sharing memory linked list, this node becomes distribution node.
In step 306, after receiving the sharing memory linked list head pointer that process 1 is transmitted, the pointer that process 2 is transmitted process 1 is converted to the memory address in the address space.Process 2 is converted to memory address in the address space with pointer data.Process 2 can be obtained the message data that is stored in the relative address by visiting described memory address, so that further process, has avoided the copy to message data like this, has saved system resource.
In step 307, sharing memory linked list in the described memory address of treatment progress 2 visits is handled accordingly to message data, after finishing processing, field result after handling is updated in this chained list, and the head pointer with this chained list sends to treatment progress 3 then.
In step 308, process 3 is obtained sharing memory linked list by the visit of memory address that the chain list index is shone upon.
In step 309, process 3 is discharged into the information node of handling in the idle node formation and goes.In the present embodiment, process 3 is last treatment progress, after message data is finished dealing with, the message queue node of having applied for can be discharged in the idle node formation and go.
Charge system is exactly a kind of typical application case of the present invention, in order to make charge system handling property is preferably arranged, and need be bundled in data and treatment progress on the same physical host to handle, and reads nearby to make things convenient for data, improves data read efficient.
Owing to can comprise tens even a hundreds of charging feature field in the charging event usually, when between process, transmitting these data,, can cause the rapid decline of systematic function if adopt the method for data copy.And each treatment progress of charge system all needs constantly to revise the data structure of charging event message and pass to next treatment progress in processing procedure, therefore for example from original charging message, identify charging feature field (as caller location, called position, Familiarity Number sign etc.), be more suitable for the method that adopts pointer of the present invention directly to quote in charge system.
In charge system as shown in Figure 5, three treatment progress have been comprised, i.e. authentication process, charging process, book keeping operation process.Three processes all are subjected to the control of dynamic programming engine, and in the present embodiment, the dynamic programming engine is the master control process of system.In actual charge system, have a plurality of treatment progress and carry out the message data processing.For convenience of description, in the present embodiment, only above-mentioned authentication, charging, book keeping operation process are described.
At first, the message data of standard ticket is input to authentication process, the feature field such as type, main body, object, conversation time started, talking position, traffic carrying capacity that has comprised traffic in the described standard ticket message data, authentication process is after handling this message data, handle the result of back gained, for example, feature field such as product IDs are added in the sharing memory linked list, and the pointer of this chained list is sent to the charging process.After the content of charging process in obtaining the memory address that the chain list index shone upon, similarly, after message data field is further handled, the result that will produce in this step (for example, fee item, expense) add in the sharing memory linked list, and the pointer of this chained list is sent to next treatment progress (process of promptly keeping accounts) handle.
Fig. 7 is the flow chart of the present invention's message delivery method in charge system.Wherein:
In step 701, shared drive district of master control process creation.The size of shared drive is calculated according to the disposal ability of system and is obtained, and also can adjust by the method for dynamic increase.
As shown in Figure 6, for fear of the memory fragmentation that when distributing less memory block, causes, the present invention has adopted memory pool technique, in memory pool, have a plurality of memory blocks, the memory cell that has predetermined length in each memory block, the field of various message datas with charge system of different length is stored in respectively in the memory block of respective length.
In step 702, master control process creation message queue.Create message queue provide can with the mutual resource of each treatment progress.Message queue is made up of a plurality of sharing memory linked list, comprising distribution node chained list and idle node chained list.
In step 703, authentication process receives the standard ticket message data from external progress, and applies for idle node from described message queue.The feature field such as type, main body, object, conversation time started, talking position, traffic carrying capacity that has comprised traffic in the described standard ticket message data.Authentication process receives the standard ticket message of transmitting from external progress by IPC such as SOCKET or FIFO mechanism, idle node of application generally includes following message: the head pointer of sharing memory linked list in the message queue node data structure from message queue; The tail pointer of sharing memory linked list; The message data field number.The idle message queue node of application is to be the message application memory block after handling.
In step 704, authentication process is to handling from the message that external progress received.Authentication process is discerned authentication process to the message data of the standard ticket imported, has generated new feature field (as product IDs) after the processing.
In step 705, the message field after authentication process will be handled (comprise and handle feature field and the original standard ticket message field of importing that the back produces) head and the tail are linked into sharing memory linked list, and the head pointer that sends this chained list then is to the charging process.Authentication process adopts IPC mechanism (for example, FIFO mechanism) that the head pointer (being the pointer of message queue node) of this chained list is sent to the charging process.The head pointer of this chained list is the head pointer of the sharing memory linked list of the authentication process idle node of applying for, owing to deposited the head pointer that points to sharing memory linked list, this node becomes distribution node.
In step 706, after receiving the sharing memory linked list head pointer that authentication process is transmitted, the charging process is converted to memory address in the address space with the pointer of authentication process transmission.The charging process is converted to memory address in the address space with pointer data, make the charging process to obtain standard ticket message data and the back field that generates of authentication process processing that is stored in the relative address by visiting described memory address, so that further process, avoided copy like this, saved system resource message data.
In step 707, the charging process is visited the sharing memory linked list in the described memory address, message data is handled accordingly, after finishing processing, the message data field of handling field result's (for example field such as fee item, expense) of back generation and originally imported from authentication process is updated to this chained list, and the head pointer with this chained list sends to the book keeping operation process then.
In step 708, the book keeping operation process is obtained sharing memory linked list by the visit of memory address that the chain list index is shone upon, and carries out respective handling, generates for example fields such as account, the amount of money, accumulation, aggregate-value.By relevant output process result is exported.
In step 709, after finishing processing, the book keeping operation process is discharged into the information node of handling in the idle node formation and goes.As last treatment progress, the book keeping operation process needs to discharge the message queue node the applied for further utilization ratio of system in the idle node formation after finishing dealing with.
Though the invention discloses the foregoing description; but specific embodiment be not be used for limiting of the present invention; any those skilled in the art; without departing from the spirit and scope of the present invention; can make some modifications and retouching, so protection scope of the present invention should be as the criterion with the scope that appending claims defines.
Claims (13)
1. the progress message transferring method based on chained list is characterized in that, comprising:
A) master control process creation shared drive district, described shared drive district is used to deposit the message data of interprocess communication;
Master control process creation message queue, described message queue comprise distribution node formation and idle node formation;
B) first treatment progress receives the message from the outside, applies for idle node from message queue, and message field is linked into sharing memory linked list from beginning to end that the head pointer that sends this chained list then is to second treatment progress;
C) after receiving the sharing memory linked list head pointer that first treatment progress transmits, second treatment progress is converted to memory address in the address space with the pointer of the first treatment progress transmission;
D) second treatment progress is visited the sharing memory linked list in the described memory address, after finishing processing the result is updated in this chained list, and the head pointer with this chained list sends to the subsequent treatment process then.
2. progress message transferring method according to claim 1 is characterized in that, described shared drive district is the memory block of each field distribution predetermined length of message.
3. progress message transferring method according to claim 1 is characterized in that, has a plurality of treatment progress that the message of outside is handled.
4. progress message transferring method according to claim 1 is characterized in that, last treatment progress in the described treatment progress is discharged into the information node of handling in the idle node formation.
5. progress message transferring method according to claim 1 is characterized in that node comprises: message queue size, message queue idle node linked list head pointer and message queue be distribution node linked list head pointer.
6. progress message transferring method according to claim 1 is characterized in that, adopts first-in first-out FIFO or other interprocess communications IPC mechanism to send the head pointer of sharing memory linked list between treatment progress.
7. progress message transferring method according to claim 1 is characterized in that, the message queue node data comprises: at least one in the head pointer of sharing memory linked list, the tail pointer of sharing memory linked list and the message data field number.
8. the charge system progress message transferring method based on chained list is characterized in that, comprising:
A) master control process creation shared drive district, described shared drive district is used to deposit the message data of interprocess communication;
Master control process creation message queue, described message queue comprise distribution node formation and idle node formation;
B) authentication process receives the message from external progress, and applies for idle node from message queue, and message field is linked into sharing memory linked list from beginning to end, and the head pointer that sends this chained list then is to the charging process;
C) after receiving the sharing memory linked list head pointer that authentication process is transmitted, the charging process is converted to memory address in this process address space with the pointer of authentication process transmission;
D) the charging process is visited the sharing memory linked list in the described memory address, after finishing processing the result after handling is updated in this chained list, and the head pointer that sends this chained list then is to the book keeping operation process;
E) book keeping operation process pointer is converted to the memory address in this process address space, visits the sharing memory linked list in the described memory address.
9. progress message transferring method according to claim 8 is characterized in that, described shared drive district is the memory block of each field distribution predetermined length of message.
10. progress message transferring method according to claim 8 is characterized in that, last treatment progress in the described treatment progress is discharged into the information node of handling in the idle node formation.
11. progress message transferring method according to claim 8 is characterized in that, node comprises: message queue size, message queue idle node linked list head pointer and message queue be distribution node linked list head pointer.
12. progress message transferring method according to claim 8 is characterized in that, adopts first-in first-out FIFO or other interprocess communications IPC mechanism to send the head pointer of sharing memory linked list between each process.
13. progress message transferring method according to claim 8 is characterized in that, the message queue node data comprises: at least one in the head pointer of sharing memory linked list, the tail pointer of sharing memory linked list and the message data field number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100074871A CN100403739C (en) | 2006-02-14 | 2006-02-14 | News transfer method based on chained list process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100074871A CN100403739C (en) | 2006-02-14 | 2006-02-14 | News transfer method based on chained list process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1859325A CN1859325A (en) | 2006-11-08 |
| CN100403739C true CN100403739C (en) | 2008-07-16 |
Family
ID=37298188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100074871A Expired - Fee Related CN100403739C (en) | 2006-02-14 | 2006-02-14 | News transfer method based on chained list process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100403739C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4060497A1 (en) * | 2021-03-19 | 2022-09-21 | Beijing Tusen Zhitu Technology Co., Ltd. | Method for inter-process communication, related computing device and storage medium |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100444120C (en) * | 2006-11-30 | 2008-12-17 | 南京联创科技股份有限公司 | Mass data transmitting method inter-pipeline multiprocess based on information queue |
| CN101141742B (en) * | 2007-10-12 | 2011-08-10 | 中兴通讯股份有限公司 | Application communication method of terminal |
| CN101458650B (en) * | 2007-12-13 | 2010-12-15 | 华为技术有限公司 | Apparatus and method for detecting internal memory transfer abnormity in multi-core system |
| CN101763289B (en) * | 2009-09-25 | 2013-11-20 | 中国人民解放军国防科学技术大学 | Message passing method based on shared memory |
| CN102906706A (en) * | 2010-05-24 | 2013-01-30 | 索尼电脑娱乐公司 | Information processing device and information processing method |
| CN102004675A (en) * | 2010-11-11 | 2011-04-06 | 福建星网锐捷网络有限公司 | Cross-process data transmission method, device and network equipment |
| CN102479106A (en) * | 2010-11-23 | 2012-05-30 | 金蝶软件（中国）有限公司 | Method and system for acquiring messages from multi-queue nodes |
| WO2013086702A1 (en) * | 2011-12-14 | 2013-06-20 | Telefonaktiebolaget L M Ericsson (Publ) | Buffer resource management method and telecommunication equipment |
| CN103197979B (en) * | 2012-01-04 | 2017-05-03 | 阿里巴巴集团控股有限公司 | Method and device for realizing data interaction access among processes |
| CN103365705B (en) * | 2012-03-27 | 2016-02-24 | 腾讯科技（深圳）有限公司 | Message queue disposal route and device |
| CN102880507A (en) * | 2012-09-12 | 2013-01-16 | 科立讯通信股份有限公司 | Method for applying and distributing chain structure message |
| CN103543988B (en) * | 2013-10-23 | 2017-05-10 | 华为终端有限公司 | Method for processing array information, method and device of controlling information to enter arrays |
| CN104536704A (en) * | 2015-01-12 | 2015-04-22 | 浪潮(北京)电子信息产业有限公司 | Dual-controller communication method, transmitting end controller and receiving end controller |
| CN105988876B (en) * | 2015-03-27 | 2019-09-17 | 杭州迪普科技股份有限公司 | Memory allocation method and device |
| CN106338931A (en) * | 2015-07-06 | 2017-01-18 | 天津九洲云物联科技有限公司 | Central control information scheduling algorithm |
| CN105045660A (en) * | 2015-07-27 | 2015-11-11 | 汉柏科技有限公司 | Dynamic resource recovery method and system |
| CN106201742B (en) * | 2016-07-04 | 2019-09-20 | 东软医疗系统股份有限公司 | A kind of shared-memory management method, apparatus and system |
| CN106681847B (en) * | 2016-12-30 | 2020-08-11 | 深圳Tcl数字技术有限公司 | Message processing method and device based on Android system |
| CN107995006A (en) * | 2017-12-01 | 2018-05-04 | 天津麒麟信息技术有限公司 | Real-time Billing System based on message trigger under a kind of cloud environment |
| CN108595282A (en) * | 2018-05-02 | 2018-09-28 | 广州市巨硅信息科技有限公司 | A kind of implementation method of high concurrent message queue |
| CN110737536B (en) * | 2019-09-19 | 2024-03-12 | 亚信创新技术(南京)有限公司 | Message storage method and message reading method based on shared memory |
| CN110750372B (en) * | 2019-10-16 | 2024-05-14 | 深圳技威时代科技有限公司 | Log system and log management method based on shared memory |
| CN111158927B (en) * | 2019-12-17 | 2023-10-24 | 科大讯飞股份有限公司 | Inter-process communication method, inter-process communication device and storage medium |
| CN111459417B (en) * | 2020-04-26 | 2023-08-18 | 中国人民解放军国防科技大学 | Non-lock transmission method and system for NVMeoF storage network |
| CN114979270B (en) * | 2022-05-25 | 2023-08-25 | 上海交通大学 | Message publishing method and system suitable for RDMA network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1490724A (en) * | 2002-10-18 | 2004-04-21 | 上海贝尔有限公司 | Virtual machine for embedded systemic software development |
| CN1501277A (en) * | 2002-11-19 | 2004-06-02 | 宝山钢铁股份有限公司 | Bus type inter-process communication method |
| WO2004061687A1 (en) * | 2002-12-19 | 2004-07-22 | Emulex Design & Manufacturing Corporation | Direct memory access controller system |
| US6847991B1 (en) * | 2000-09-06 | 2005-01-25 | Cisco Technology, Inc. | Data communication among processes of a network component |
| US20050044551A1 (en) * | 2003-08-19 | 2005-02-24 | Sodhi Ajit S. | System and method for shared memory based IPC queue template having event based notification |
| CN1713151A (en) * | 2004-06-22 | 2005-12-28 | 国际商业机器公司 | Method and system for facilitating communication within shared memory environments using lock-free queues |
-
2006
- 2006-02-14 CN CNB2006100074871A patent/CN100403739C/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6847991B1 (en) * | 2000-09-06 | 2005-01-25 | Cisco Technology, Inc. | Data communication among processes of a network component |
| CN1490724A (en) * | 2002-10-18 | 2004-04-21 | 上海贝尔有限公司 | Virtual machine for embedded systemic software development |
| CN1501277A (en) * | 2002-11-19 | 2004-06-02 | 宝山钢铁股份有限公司 | Bus type inter-process communication method |
| WO2004061687A1 (en) * | 2002-12-19 | 2004-07-22 | Emulex Design & Manufacturing Corporation | Direct memory access controller system |
| US20050044551A1 (en) * | 2003-08-19 | 2005-02-24 | Sodhi Ajit S. | System and method for shared memory based IPC queue template having event based notification |
| CN1713151A (en) * | 2004-06-22 | 2005-12-28 | 国际商业机器公司 | Method and system for facilitating communication within shared memory environments using lock-free queues |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4060497A1 (en) * | 2021-03-19 | 2022-09-21 | Beijing Tusen Zhitu Technology Co., Ltd. | Method for inter-process communication, related computing device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1859325A (en) | 2006-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100403739C (en) | News transfer method based on chained list process | |
| US8930593B2 (en) | Method for setting parameters and determining latency in a chained device system | |
| CN101636715B (en) | Instruction communication method and device for multi-processor system | |
| CN102209104B (en) | Reducing packet size in communication protocol | |
| CN102761489B (en) | Inter-core communication method realizing data packet zero-copying based on pipelining mode | |
| CN101848203B (en) | Apparatus and method for supporting connection establishment in an offload of network protocol processing | |
| CN110191194B (en) | RDMA (remote direct memory Access) network-based distributed file system data transmission method and system | |
| CN102255794B (en) | Remote message transmit-receive handling capacity and waiting time shortening system and method | |
| US7694310B2 (en) | Method for implementing MPI-2 one sided communication | |
| US20120066692A1 (en) | Iteratively processing segments by concurrently transmitting to, processing by, and receiving from partnered process | |
| ITRM20120094A1 (en) | NETWORK INTERFACE CARD FOR PARALLEL CALCULATION NETWORK KNOT ON GPU, AND RELATIVE INTERNODAL COMMUNICATION METHOD | |
| CN105978985A (en) | Memory management method of user-state RPC over RDMA | |
| US7457845B2 (en) | Method and system for TCP/IP using generic buffers for non-posting TCP applications | |
| CN1980198A (en) | Method and apparatus for striping message payload data over a network | |
| CN112698959A (en) | Multi-core communication method and device | |
| CN111459417A (en) | NVMeoF storage network-oriented lock-free transmission method and system | |
| WO2023098050A1 (en) | Remote data access method and apparatus | |
| CN103324599A (en) | Inter-processor communication method and system on chip | |
| US7552232B2 (en) | Speculative method and system for rapid data communications | |
| CN115129621A (en) | Memory management method, device, medium and memory management module | |
| CN105320676A (en) | Customer data query service method and device | |
| US10489322B2 (en) | Apparatus and method to improve performance in DMA transfer of data | |
| CN113127139A (en) | Memory allocation method and device based on data plane development kit DPDK | |
| CN116775522A (en) | Data processing method based on network equipment and network equipment | |
| CN104199781A (en) | Memory fragment allocation method and device based on shared memory |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080716 Termination date: 20130214 |