CN106844242B - A kind of method for interchanging data and system - Google Patents
A kind of method for interchanging data and system Download PDFInfo
- Publication number
- CN106844242B CN106844242B CN201611258888.4A CN201611258888A CN106844242B CN 106844242 B CN106844242 B CN 106844242B CN 201611258888 A CN201611258888 A CN 201611258888A CN 106844242 B CN106844242 B CN 106844242B
- Authority
- CN
- China
- Prior art keywords
- data
- barrier
- node
- consumption
- address bus
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1605—Handling requests for interconnection or transfer for access to memory bus based on arbitration
- G06F13/161—Handling requests for interconnection or transfer for access to memory bus based on arbitration with latency improvement
- G06F13/1615—Handling requests for interconnection or transfer for access to memory bus based on arbitration with latency improvement using a concurrent pipeline structrure
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to method for interchanging data and system.This method comprises: the data in Xiang Wusuo data/address bus write-in data generate node and establish the first barrier between the data consumption node for reading data without lock data/address bus；The second barrier is being established from the data consumption node without lock data/address bus reading data and to the shown data generation node without lock data/address bus write-in data；It monitors first barrier and whether second barrier changes；And when first barrier changes, the data consumption node reads data, and when second barrier changes, the data generate node and generate data；Wherein the data generate node, the first barrier, the data consumption node and second barrier and form circulation.
Description
Technical field
The present invention relates generally to mobile services to support field, more particularly, to a kind of method for interchanging data and system.
Background technique
Currently, concurrent data exchange uses reading substantially between each business module in Mobile Service Support System field
It writes lock or compares and exchange the mode of (CAS) operation to carry out con current control.In multiple industry for carrying out business datum exchange
In module of being engaged in, the business datum exchanged write-in will be needed to be used in the storage medium of data exchange first by data generating module
(file, database etc.) then reads data from storage medium by data consumption module.When same part data needs are more
When a data consumption module accesses, the quantity according to consumption module is needed, data are replicated more parts, disappeared respectively to different data
Take module use.Data consumption module is after consumer behavior completion, voluntarily deletion of data content, to guarantee that data have and only
One-time-consumption behavior occurs.
In above process, it generally needs that the data in storage medium are compared and exchanged plus Read-Write Locks, or use
(CAS) operation carries out con current control, to prevent the dirty reading of exchange data, misread, i.e., adds between " producer " and " consumer "
Read-Write Locks carry out CAS operation.Due to writing exclusivity, the exclusiveness of lock, read lock and the alternative and CAS operation of writing lock
Parallel work-flow can be all converted into serial operation by serial property, to cause very big expense to system, significantly limited and be based on
Multiple business modules of above-mentioned data exchange ways restrict business support system to the concurrency performance of same a business datum read-write
The business throughput of system.It, not only can be due in addition, replicate more parts of way by data consumer quantity for same part data
Data replicate and increase system response time, and waste data space, cause the hardware costs such as cpu resource, storage resource
Increase.
Summary of the invention
According to an aspect of the present invention, the present invention provides a kind of method for interchanging data.This method comprises: in Xiang Wusuo
The data of data/address bus write-in data generate node and establish between the data consumption node for reading data without lock data/address bus
First barrier；It is write from the data consumption node without lock data/address bus reading data and to shown without lock data/address bus
The data for entering data generate node and establish the second barrier；It monitors first barrier and whether second barrier changes；
And when first barrier changes, the data consumption node reads data, when second barrier changes
When, the data generate node and generate data；Wherein the data generate node, the first barrier, the data consumption node and
Second barrier forms circulation.
According to another aspect of the present invention, the present invention provides a kind of data exchange systems.The system includes: the first screen
Hinder setup module, the first barrier setup module be configured to Xiang Wusuo data/address bus be written data data generate node and
The first barrier is established between the data consumption node for reading data without lock data/address bus；Second barrier setup module, this second
Barrier setup module is configured as without lock data/address bus reading the data consumption node of data and to shown nothing from described
The data of lock data/address bus write-in data generate node and establish the second barrier；Monitoring modular, the monitoring modular are configured as
It monitors first barrier and whether second barrier changes；And ruling module, the ruling module are configured as when described
When first barrier changes, determine that the data consumption node reads data, when second barrier changes, determines
The data generate node and generate data；Wherein the data generate node, the first barrier, the data consumption node and described
Second barrier forms circulation.
According to another aspect of the present invention, the present invention also provides a kind of machine readable storage devices comprising machine
Readable instruction realizes the above method or system when the machine readable instructions are performed.
Detailed description of the invention
The present invention may be better understood from the description with reference to the accompanying drawing to a specific embodiment of the invention,
In:
Fig. 1 shows the structural schematic diagram according to an embodiment of the present invention without lock data/address bus；
Fig. 2 shows the processes of the method according to an embodiment of the present invention that data exchange is carried out based on Wireless digital bus
Figure；
The structure that Fig. 3 shows the system according to an embodiment of the present invention that data exchange is carried out based on Wireless digital bus is shown
It is intended to；
Fig. 4 shows the working principle diagram of the method for data exchange according to an embodiment of the present invention；
Fig. 5 shows the timing diagram of the method for the serial data exchange of data consumption node according to an embodiment of the present invention；
Fig. 6 shows the timing diagram of the method for the data exchange of data consumption nodal parallel according to an embodiment of the present invention.
Specific embodiment
The feature and exemplary embodiment of various aspects of the invention is described more fully below.In following detailed description
In, many details are proposed, in order to provide complete understanding of the present invention.But to those skilled in the art
It will be apparent that the present invention can be implemented in the case where not needing some details in these details.Below to implementation
The description of example is used for the purpose of providing by showing example of the invention and better understanding of the invention.The present invention never limits
In any concrete configuration set forth below and algorithm, but cover under the premise of without departing from the spirit of the present invention element,
Any modification, replacement and the improvement of component and algorithm.In the the accompanying drawings and the following description, well known structure and skill is not shown
Art is unnecessary fuzzy to avoid causing the present invention.
Fig. 1 shows the structural schematic diagram without lock data/address bus.As shown in Figure 1, the memory without lock data/address bus can be with
It is divided into N number of unit.The data managed without lock data/address bus may be present in system shared drive or process is privately owned
In memory.The data without lock data/address bus generate node (producer) and data consumption node (consumer) uses login mechanism
Notify this without lock data/address bus.It is write in general, only allowing to register a data without lock data/address bus and generating node (producer)
Enter operation, but allows to register multiple data consumption nodes (consumer) and be read.This without lock data/address bus in order, according to
The secondary unit (1,2,3 ... N) that equal part is recycled: data generate node (producer) and successively write number toward unit (1,2,3 ... N)
According to data consumption node (consumer) successively reads data from unit (1,2,3 ... N).In general, only when data are by all data
After consuming node (consumer) reading (consumption), unit (1,2,3 ... N) just can generate node (producer) by data and reuse
(circulation covering), i.e., be written data again.
Fig. 2 shows the processes of the method according to an embodiment of the present invention that data exchange is carried out based on Wireless digital bus
Figure.The method shown in Fig. 2 for carrying out data exchange based on Wireless digital bus includes: S210 in Xiang Wusuo data/address bus write-in number
According to data generate node and from without lock data/address bus read data data consumption node between establish the first barrier；S220 exists
The data consumption node of data is read and to shown without the institute for locking data/address bus write-in data without lock data/address bus from described
It states data generation node and establishes the second barrier；S230 monitors first barrier and whether second barrier changes；And
For S240 when first barrier changes, the data consumption node reads data, and S250 occurs when second barrier
When variation, the data generate node and generate data.
Fig. 3 shows the knot of the system 300 according to an embodiment of the present invention that data exchange is carried out based on Wireless digital bus
Structure schematic diagram, the system 300 include: the first barrier setup module 310, the first barrier setup module 310 be configured to
Data without lock data/address bus write-in data generate node and read between the data consumption node of data from without lock data/address bus
Establish the first barrier；Second barrier setup module 320, the second barrier setup module 320 be configured as from it is described without lock number
The data consumption node of data is read according to bus and generates section to the shown data without lock data/address bus write-in data
Point establishes the second barrier；Monitoring modular 330, the monitoring modular 330 are configured as monitoring first barrier and second screen
Whether barrier changes；And ruling module 340, the ruling module 340 are configured as when first barrier changes, and are determined
The data consumption node reads data, when second barrier changes, determines that the data generate node and generate number
According to.
Fig. 4 shows the working principle diagram of the method for data exchange according to an embodiment of the present invention.In Fig. 4, " production
Person " is for without lock data/address bus, is write-in data side；" consumer " is
Data side is read, wherein " consumer " can also modify to the data without data with existing unit in lock data/address bus, but never
Data cell can be increased newly；" Wr_sequnce_No ", " Rd_sequnce_No " be a computer " word ", be " producer " and
" barrier " between " consumer " guarantees exact sequence operation between the two.This during read-write, is utilized without lock data/address bus
Modem computer systems carry out atomic operation and the extremely low characteristic of expense to " word ", in " producer " and " consumer " to " barrier "
When concurrent operations, not plus Read-Write Locks are also operated without " CAS ", by the natural atomicity of computer system " word arithmetic ", are guaranteed
Sequence between data exchange concurrent operations.In Fig. 4, between each pair of " producer " and " consumer ", have and only one
" barrier ".In one embodiment, it may be defined as sequence between consumer to read, such as " consumer 1 " and " consumer 2 ".One
It may be defined as reading parallel in a embodiment, between consumer, such as " consumer 1 " and " consumer 3 ".
In one embodiment, upon power-up of the system, barrier " Wr_ is established between " producer " and " consumer 1 "
sequnce_No".After " producer " generates a data, by the barrier stepping one, " consumer 1 " discovery " Wr_
When sequnce_No " changes, the data of " producer " generation can be read safely, this process is without Read-Write Locks or CAS operation, effectively
Promote concurrency performance.
In another embodiment, upon power-up of the system, barrier " Rd_ is established between " consumer 1 " and " consumer 2 "
sequnce_No1".After " consumer 1 " has consumed a data, by the barrier stepping one, " consumer 2 " discovery " Rd_
When sequnce_No1 " changes, the data of " consumer 1 " post-consumer can be read safely, this process without Read-Write Locks or CAS operation,
Effectively promote concurrency performance.Upon power-up of the system, barrier " Rd_sequnce_ is established between " consumer 2 " and " producer "
No2".After " consumer 2 " have consumed a data, by the barrier stepping one, after the full cyclic annular queue of " producer " sequential write,
When generating new data again, by the step value of inspection " Rd_sequnce_No2 ", judge data whether by last
Position consumer spending is complete.If consumption is complete, the data storage cell of post-consumer can be covered, be recycled, if not consumed,
" producer " waits until data have been consumed, this process is without Read-Write Locks or CAS operation, effectively promotion concurrency performance.
For " consumer 2 ", " consumer 1 " can be regarded as " producer " of " consumer 2 ".As with " consumer
1 " " consumer 3 " read parallel, the barrier between " producer " are also " Wr_sequnce_No ", process of consumption, with
" consumer 1 " is similar.
Fig. 5 shows the timing diagram of the method for the serial data exchange of data consumption node according to an embodiment of the present invention.
In the timing diagram of Fig. 5, step 1, " producer " generates data, and queue is written；
Step 2, the barrier between " producer " stepping and " consumer 1 ": Wr_sequnce_No；
Step 3, whether " consumer 1 " monitoring Wr_sequnce_No changes, if changing, carries out step 4,
Otherwise continue to monitor；
Step 4, " consumer 1 " starts consumption data；
Step 5, after " consumer 1 " has consumed data, barrier between stepping and " consumer 2 ": Rd_sequnce_No1；
Step 6, whether " consumer 2 " monitoring Rd_sequnce_No1 changes, if changing, carries out step 7,
Otherwise continue to monitor；
Step 7, " consumer 2 " start consumption data；
Step 8, after " consumer 2 " have consumed data, barrier between stepping and " producer ": Rd_sequnce_No2；
Step 9, after the full cyclic annular queue of " producer " sequential write, used storage unit is being covered to generate new number
According to before, by judgement and last consumer --- the step value of the barrier Rd_sequnce_No2 between " consumer 2 ",
Determine whether last consumer has consumed data；
Step 10, if " consumer 2 " have not consumed data also, " producer " waits until data are disappeared by " consumer 2 "
Take；
Step 11, " producer " reuses data storage cell, and data storage cell has been consumed in covering, and is produced new
Data.
Fig. 6 shows the timing diagram of the method for the data exchange of data consumption nodal parallel according to an embodiment of the present invention.
Step 1, " producer " generates data, and queue is written；
Step 2, the barrier between " producer " stepping and " consumer 1 ": Wr_sequnce_No；
Step 3, whether " consumer 1 " monitoring Wr_sequnce_No changes, if changing, carries out step 4,
Otherwise continue to monitor；
Step 4, " consumer 1 " starts consumption data；
Step 5, after " consumer 1 " has consumed data, barrier between stepping and " producer ": Rd_sequnce_No1；
Step 6, whether " consumer 3 " monitoring Wr_sequnce_No variation occurs, if changing, carries out step 4,
Otherwise continue monitoring (this step can be parallel with step 3)；
Step 7, " consumer 3 " start consumption data (this step can be parallel with step 4)；
Step 8, after " consumer 3 " have consumed data, barrier between stepping and " producer ": Rd_sequnce_No3
(this step can be parallel with step 5)；
Step 9,10 are needing to cover used storage unit after the full cyclic annular queue of " producer " sequential write to produce
Before raw new data, pass through judgement and all consumers --- the barrier Rd_ between " consumer 1 ", " consumer 3 "
The stepped parameter of sequnce_No1, Rd_sequnce_No3, determine whether all consumers have consumed data；
Step 11, if there is any consumer not consume data also, " producer " waits until data are by all consumption
Person has consumed；
Step 12, " producer " reuses data storage cell, and data storage cell has been consumed in covering, and is produced new
Data.
Ticket in Billing (charging) system, after preprocessed resume module, it usually needs give subsequent module (example
Such as, wholesale price module and account discount module) it is handled.It is using disk file between these three modules in existing system
Data interaction is carried out, to prevent wholesale price module and account discount module while handling same file, leads to data in file
Misread, dirty reading the problem of, during processing, preprocessing module can first by oneself processing complete ticket, be written to interim text
It in part, then timing or quantitatively renames into official document, gives wholesale price resume module；Wholesale price module is also timing or quantifies, and is generated just
Formula CDR file gives account discount resume module.However, the presence and intermodular data circulation due to file read-write lock stop
(as generated official document at regular time and quantity), this tupe typically result in that system process performance is high, user bill processing is prolonged
When.In addition, when most field contents in the CDR file needed for the wholesale price and account discount module are identical, the processing
Mode causes to store redundancy.
In contrast, after the component for using the data/address bus of the application to exchange as intermodular data, preprocessing module is
As " producer ", wholesale price module and account discount module i.e. as " consumer 1 " and " consumer 2 " serially consumed, due to
Storage becomes the promotion of memory and bring readwrite performance, and the promotion of bring concurrency performance without lock from disk, guarantees
The reduction of the real-time and memory space redundancy of user bill processing.
In one embodiment, a kind of machine readable storage device is provided comprising machine readable instructions, when the machine
Readable instruction is performed, and realizes the above method or device.
The embodiment and implementation of systems and methods described herein may include various operations, these operations can quilt
It is implemented in the machine-executable instruction executed by computer system.Computer system may include one or more general-purpose computations
Machine or special purpose computer (or other electronic equipments).Computer system may include hardware component, these hardware components include using
In execute operation certain logic or may include hardware, software and/or firmware combination.
Various technologies or its some aspects or part can be using the form of program code (that is, instruction), described program generations
Code is implemented in that such as floppy disk, CD-ROM, hard-drive, magnetic or optical card, solid-state memory device, non-transient computer are readable deposits
Store up medium or any other machine-readable storage media, wherein when program code is loaded into machine (for example, computer) simultaneously
When being executable by a machine, which becomes the device for implementing various technologies.It is run on programmable computers in program code
When, calculate equipment may include processor, can be by storaging medium (including volatile and non-volatile memory that processor is read
And/or memory element), at least one input equipment and at least one output equipment.Volatile and non-volatile memory and/
Or memory element can be RAM, EPROM, flash drive, CD-ROM driver, magnetic hard drives or for storing electronic data
Other media.
It should be understood that many functional units described in this specification can be implemented as one or more components,
Component is for especially emphasizing that it realizes the term of independence.For example, component may be implemented as hardware circuit, the hardware circuit
Super large-scale integration (VLSI) circuit or gate array or such as logic chip, transistor or other points including customization
The ready-made semiconductor of vertical component etc.Component can also be implemented in programmable hardware device, for example, field-programmable gate array
Column, programmable logic array, programmable logic device etc..
Component can also be implemented in software to be executed by various types of processors.The executable code identified
Component for example may include one or more physical or logic blocks of computer instruction, such as can be organized as object, process
Or function.It however, the executable code of the component identified is physically not necessarily together, but may include being stored in
Isolated instruction in different location may include the component when the instruction of these separation is logically combined together
And realize the target of the component defined.
In fact, the component of executable code can be single instruction or multiple instruction, and if can even be distributed in
Dry between different code snippets, different programs and several memory devices.Similarly, herein, operation data
It can be identified and be shown in component, and any suitable mode can be used to realize and be arranged in any type
Data structure in.Operation data can be set as individual data collection, or can be distributed in different positions (including different
Store equipment), and at least partly can only exist as system or the electronic signal of network.Component can be it is passive or
Active, the agency of desired function is executed including that can operate.
Several aspects of described embodiment will be illustrated as software module or component.As it is used herein,
Software module or component may include any type of computer instruction or computer executable generation in memory devices
Code.Software module for example may include one or more computer instruction physical blocks or logical block, can be arranged to perform one
A or multiple tasks or routine, programs, objects, component, the data structure etc. for realizing specific data type.It should be understood that generation
For software or in addition to software, software module can be implemented in hardware and/or firmware.One or more function described herein
Energy module can be separated into submodule and/or be combined into single or a small amount of module.
In certain embodiments, specific software module may include that the different location for being stored in memory devices, difference are deposited
Described functions of modules is realized in isolated instruction in storage device or different computers, these instructions together.In fact, mould
Block may include single instruction or multiple instruction, and can be distributed in code snippet several different, different programs with
And between several memory devices.Some embodiments can be incorporated in distributed computing environment, wherein by passing through communication network
The remote processing devices of network link execute task.In a distributed computing environment, software module can be located locally and/or remotely
In memory storage device.In addition, the data for bundling or presenting together in data-base recording may reside within same memory
In equipment or in several memory devices, and it can be linked together in the record field in the database of network.
Although foregoing teachings are described in detail for purposes of clarity, but it will be apparent that without departing substantially from its principle
In the case of can make certain changes and modification.It should be noted that there are many realization process described hereins and dresses
The alternative set.Therefore, the embodiments herein is illustrative rather than restrictive, and the present invention is not limited to given herein
Details, but can modify in scope of the appended claims and in being equal, also, fall into the meaning of claim and wait
Whole changes in the range of jljl are to all be included among the scope of the present invention.
Claims (19)
1. a kind of method for interchanging data characterized by comprising
Node is generated in the data of Xiang Wusuo data/address bus write-in data and from without the data consumption for locking data/address bus reading data
The first barrier is established between node, first barrier generates after node generates data for the data and carries out stepping mark；
Node is generated in the data consumption node and the data and establishes the second barrier, and second barrier is used for；
It monitors first barrier and whether second barrier changes；And
When first barrier changes, the data consumption node reads data, when second barrier changes
When, the data generate node and generate data.
2. method for interchanging data according to claim 1, wherein first barrier and second barrier are a meters
Calculation machine word.
3. method for interchanging data according to claim 1, wherein described only allow to register a data without lock data/address bus
Generate node.
4. method for interchanging data according to claim 1, allows to register multiple data wherein described without lock data/address bus and disappear
Take node.
5. method for interchanging data according to claim 4, which is characterized in that the multiple data consumption node is that sequence is read
It takes.
6. method for interchanging data according to claim 5, which is characterized in that two in the multiple data consumption node
Barrier is established between a adjacent data consumption node, wherein preceding data consumption node can be regarded as posterior data consumption
The data of node produce node.
7. method for interchanging data according to claim 4, which is characterized in that the multiple data consumption node is to read parallel
It takes.
8. method for interchanging data according to claim 7, which is characterized in that generate node and the multiple in the data
Barrier is established respectively between data consumption node.
9. method for interchanging data according to claim 1 to 8, which is characterized in that when the data generate node
After generating a data, described the stepping of first barrier one；After the data consumption node reads a data, described second
Barrier stepping one.
10. a kind of data exchange system characterized by comprising
First barrier setup module, the first barrier setup module are configured as that the data of data are written in Xiang Wusuo data/address bus
It generates node and establishes the first barrier between the data consumption node for reading data without lock data/address bus；
Second barrier setup module, the second barrier setup module are configured as reading data without lock data/address bus from described
The data consumption node and to it is shown without lock data/address bus write-in data the data generate node establish the second barrier；
Monitoring modular, which is configured as monitoring first barrier and whether second barrier changes；And
Decide that module, the ruling module are configured as determining the data consumption node when first barrier changes
Data are read, when second barrier changes, determine that the data generate node and generate data；
Wherein the data generate node, the first barrier, the data consumption node and second barrier and form circulation.
11. data exchange system according to claim 10, wherein first barrier and second barrier are one
Computer word.
12. data exchange system according to claim 10, wherein described only allow to register a number without lock data/address bus
According to generation node.
13. data exchange system according to claim 10, wherein described allow to register multiple data without lock data/address bus
Consume node.
14. data exchange system according to claim 13, which is characterized in that the multiple data consumption node is sequence
It reads.
15. data exchange system according to claim 14, which is characterized in that in the multiple data consumption node
Barrier is established between two adjacent data consumption nodes, is disappeared wherein preceding data consumption node can be regarded as posterior data
Take the data production node of node.
16. data exchange system according to claim 13, which is characterized in that the multiple data consumption node is parallel
It reads.
17. data exchange system according to claim 16, which is characterized in that generate node and described more in the data
Barrier is established respectively between a data consumption node.
18. data exchange system described in any one of 0-17 according to claim 1, which is characterized in that when the data generate
After node generates a data, described the stepping of first barrier one；It is described after the data consumption node reads a data
Second barrier stepping one.
19. a kind of machine readable storage device comprising machine readable instructions, it is real when the machine readable instructions are performed
The now method or system as described in any claim in claim 1-18.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611258888.4A CN106844242B (en) | 2016-12-30 | 2016-12-30 | A kind of method for interchanging data and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611258888.4A CN106844242B (en) | 2016-12-30 | 2016-12-30 | A kind of method for interchanging data and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106844242A CN106844242A (en) | 2017-06-13 |
| CN106844242B true CN106844242B (en) | 2019-07-05 |
Family
ID=59114311
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611258888.4A Active CN106844242B (en) | 2016-12-30 | 2016-12-30 | A kind of method for interchanging data and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106844242B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107656804B (en) * | 2017-09-22 | 2019-03-01 | 深圳金融电子结算中心有限公司 | Process pool system and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218176A (en) * | 2013-04-02 | 2013-07-24 | 中国科学院信息工程研究所 | Data processing method and device |
| CN103391256A (en) * | 2013-07-25 | 2013-11-13 | 武汉邮电科学研究院 | Base station user plane data processing and optimizing method based on Linux system |
| CN104077113A (en) * | 2014-07-10 | 2014-10-01 | 中船重工（武汉）凌久电子有限责任公司 | Method for achieving unlocked concurrence message processing mechanism |
| CN104978321A (en) * | 2014-04-02 | 2015-10-14 | 阿里巴巴集团控股有限公司 | Method and device for constructing data queue, method for inserting object into data queue and method for consuming object from data queue |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8868604B2 (en) * | 2012-09-26 | 2014-10-21 | Oracle International Corporation | Methods and apparatus for implementing Semi-distributed Lock Management |
| US9460143B2 (en) * | 2014-02-14 | 2016-10-04 | Oracle International Corporation | Methods, systems, and computer readable media for a multi-view data construct for lock-free operations and direct access |
-
2016
- 2016-12-30 CN CN201611258888.4A patent/CN106844242B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218176A (en) * | 2013-04-02 | 2013-07-24 | 中国科学院信息工程研究所 | Data processing method and device |
| CN103391256A (en) * | 2013-07-25 | 2013-11-13 | 武汉邮电科学研究院 | Base station user plane data processing and optimizing method based on Linux system |
| CN104978321A (en) * | 2014-04-02 | 2015-10-14 | 阿里巴巴集团控股有限公司 | Method and device for constructing data queue, method for inserting object into data queue and method for consuming object from data queue |
| CN104077113A (en) * | 2014-07-10 | 2014-10-01 | 中船重工（武汉）凌久电子有限责任公司 | Method for achieving unlocked concurrence message processing mechanism |
Non-Patent Citations (2)
| Title |
|---|
| CAS-based lock-free algorithm for shared deques;Maged M. Michael;《 Euro-Par 2003 Parallel Processing 》;20030829;全文 |
| 面向并行安全网关流水线模型的无锁队列算法;高志民，姚崎;《北京交通大学学报》;20101031;第34卷(第5期);全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106844242A (en) | 2017-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210374610A1 (en) | Efficient duplicate detection for machine learning data sets | |
| US9619430B2 (en) | Active non-volatile memory post-processing | |
| US20200050968A1 (en) | Interactive interfaces for machine learning model evaluations | |
| CA2953826C (en) | Machine learning service | |
| CN103177057B (en) | Many accounting methods for internal memory column storage database | |
| US9122676B2 (en) | License reconciliation with multiple license types and restrictions | |
| CN111339073A (en) | Real-time data processing method and device, electronic equipment and readable storage medium | |
| CN107644286A (en) | Workflow processing method and device | |
| CN101587491A (en) | Hybrid database system using runtime reconfigurable hardware | |
| EP2763055B1 (en) | A telecommunication method and mobile telecommunication device for providing data to a mobile application | |
| CN104067282A (en) | Counter operation in a state machine lattice | |
| CN103544153A (en) | Data updating method and system based on database | |
| US10656964B2 (en) | Dynamic parallelization of a calculation process | |
| US20140122160A1 (en) | Optimized License Procurement | |
| CN109978392B (en) | Agile software development management method and device, electronic equipment and storage medium | |
| CN112579586A (en) | Data processing method, device, equipment and storage medium | |
| JP2022547673A (en) | DATA PROCESSING METHOD AND RELATED DEVICE, AND COMPUTER PROGRAM | |
| Burgueño et al. | On the concurrent execution of model transformations with linda | |
| CN110727664A (en) | Method and device for executing target operation on public cloud data | |
| CN113849503A (en) | Open big data processing system, method and medium | |
| CN106844242B (en) | A kind of method for interchanging data and system | |
| CN105573763A (en) | Embedded system modeling method supporting RTOS | |
| CN115329011A (en) | Data model construction method, data query method, data model construction device and data query device, and storage medium | |
| Xiao et al. | ChoroWare: a software toolkit for choropleth map classification | |
| Li et al. | An application-oblivious memory scheduling system for DNN accelerators |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |