CN109597679A - A kind of distributed computing processing method - Google Patents
A kind of distributed computing processing method Download PDFInfo
- Publication number
- CN109597679A CN109597679A CN201811468753.XA CN201811468753A CN109597679A CN 109597679 A CN109597679 A CN 109597679A CN 201811468753 A CN201811468753 A CN 201811468753A CN 109597679 A CN109597679 A CN 109597679A
- Authority
- CN
- China
- Prior art keywords
- slave
- host
- distributed computing
- task
- processing method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/465—Distributed object oriented systems
Abstract
The invention discloses a kind of distributed computing processing methods, comprising the following steps: a general assignment is split into the small task of N number of same type by host；Host acquisition can use slave list；Small task is distributed into multiple slaves and executes calculating；The task data of slave receiving host distribution completes calculating task according to data processor, and feeds back calculated result to host；The calculated result that host is fed back according to each slave summarizes the calculating for completing general assignment.The method of the present invention supports a plurality of types of computers to participate in setting up, and can support to work under the network environment of various complexity, can according to need and in real time, easily and flexibly constructs, all kinds of idle computer resources can be enable to effectively integrate and make full use of；The method of the present invention facilitates the suitable application area for expanding distributed computing using simple, widely applicable, is conducive to the promotion and popularization for accelerating distributed computing technology.
Description
Technical field
The present invention relates to computer technology more particularly to a kind of distributed computing processing methods.
Background technique
With the continuous development of computer technology and the increasingly increase of calculation scale, distributed computing technology is just increasingly
It is valued by people.The method of distributed computing is total by needing very huge computing capability that could handle for one
Task splits into many small tasks, and each small task of fractionation is then distributed to many computer nodes and is located in parallel
Reason, finally integrates the calculated result of each node to obtain final task result.Since distributed computing is with superpower
Computing capability, high speed memory bandwidth, support large-scale data grade parallel Programming the features such as, in extensive, high-performance
Calculating field obtained preferably applying, and embody inexpensive, it is efficient, easy to maintain the advantages that.
Currently, distributed computing focuses mostly in some specific computer clusters, such as a certain research institution, group's public affairs
The groups such as department, school, each calculate node usually pass through LAN connection, and still few in cross-cutting, across group application study, make
Distributed computing technology use and popularize it is far from enough.It is various as the high speed of information network worldwide is universal
Soft and hardware resource nearly all realizes seamless connection with Internet, therefore, research and performance distributed computing technology, with energy
All kinds of idle computer resources around are enough made full use of, realization is cheap whenever and wherever possible, efficiently handles all kinds of large-scale tasks, makes
Distributed computing can preferably serve people's lives, to the science for realizing resource consolidation, energy conservation and environmental protection and the quickening mankind
Process has great significance.
Summary of the invention
It is an object of the invention to propose a kind of distributed computing processing method, the network rings in various complexity can be supported
It works under border, it is adaptable, widely applicable, to expand the applicable surface of distributed computing well, be conducive to accelerate distribution
The promotion and popularization of formula computing technique.
To realize the above-mentioned technical purpose, the technical solution adopted is as follows:
A kind of distributed computing processing method, comprising the following steps:
Step (1): host receives general assignment, and general assignment is split into N number of small task, mutually indepedent between the small task；
Step (2): host acquisition can use slave list；
Step (3): N number of small task is distributed to multiple available slaves and executes calculating by host；
Step (4): the task data that can be distributed with slave receiving host completes calculating task according to data processor, and anti-
Calculated result is presented to host；
Step (5): the calculated result that host is fed back according to each slave summarizes the calculating for completing general assignment.
Further, the host is desktop computer, laptop, tablet computer, net book, super basis, intelligent hand
One of machine, server.
Further, the slave is desktop computer, laptop, tablet computer, net book, super basis, intelligent hand
One of machine, server.
Further, the step (2) specifically includes:
Step (21): host finds slave；
Step (22): host judges whether slave can be used.
Further, it includes being found by aggressive mode and/or being found by Passive Mode that the host, which finds slave,；Institute
It states aggressive mode and looks for slave for main owner is dynamic；The Passive Mode is that host passively waits slave to connect oneself.
Further, the aggressive mode is scanning in local area network, bluetooth scans, nfc card is read, in remote Slave mode
It is one or more；The Passive Mode is one of LAN Access, Bluetooth broadcast, nfc card simulation, C/S model or more
Kind.
Further, the sequencing that each aggressive mode or Passive Mode enable is respectively as follows: scanning, local area network in local area network
Monitoring, C/S model, bluetooth scanning, Bluetooth broadcast, nfc card reading/nfc card simulation, remote Slave mode.
Further, the host judges whether slave is available and specifically includes following two aspect:
1) can connectivity judgement, i.e., whether host and slave can connect unobstructed；
2) whether slave has being currently running for task.
Further, the source mode of the data processor is one of following three kinds of modes:
Host is distributed together with task data to slave；
It is installed in advance on slave；
It is downloaded on host by slave.
The invention has the following advantages:
(1) a kind of distributed computing processing method of the invention, by the way that general assignment is split as multiple small tasks and is distributed to more
A different slave executes, and greatly reduces the consumption that the hardware and time resource of general assignment are completed by single computer, together
When can shorten time of completion task, improve task execution efficiency.
(2) a kind of distributed computing processing method of the invention supports a plurality of types of computers to participate in setting up, and energy
It is enough to support to work under the network environment of various complexity, it can according to need and whenever and wherever possible, flexibly and easily construct and use, from
And all kinds of idle computer resources can be enable to effectively integrate and utilize in real time, realize the rational allocation and Maximum Value of resource
Change.
(3) the method for the present invention is easily achieved, and using simple, adaptable, widely applicable and low in cost, high-efficient, is had
The suitable application area for helping expand distributed computing is conducive to the promotion and popularization for accelerating distributed computing technology.
Detailed description of the invention
Fig. 1 is distributed computing processing method flow chart of the present invention.
Fig. 2 is the model selection flow chart that host of the present invention finds slave.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.
The purpose of the present invention is to provide a kind of distributed computing processing method, the network rings in various complexity can be supported
It works under border, it is adaptable, widely applicable, to expand the applicable surface of distributed computing well, be conducive to accelerate distribution
The promotion and popularization of formula computing technique.
Fig. 1 is distributed computing processing method flow chart of the present invention.Below in conjunction with Fig. 1, distribution of the invention is discussed in detail
Steps are as follows for computation processing method:
Step (1): host receives general assignment, general assignment is split into N number of small task, N number of small task is the small task of same type
And it is mutually indepedent.The general assignment that the present invention program is directed to need to meet the condition of following three aspects:
1) the small task of multiple same types can be split as；
2) mutually indepedent between the small task after splitting, it cannot interdepend；
It 3) can be by each small task of processing as a result, to be finally completed general assignment.
Such as: summation work is carried out to a series of digital (99,33,660,89,106,108,200,3000,5006,9000)
For a general assignment, this general assignment is split as following 5 small tasks to complete, each small task is 2 groups of number summations:
(99,33), (660,89), (106,108), (200,3000), (5006,9000)；
Every 2 groups of numbers summation is the small task of same type above, and each small task read group total is mutually indepedent, does not interdepend respectively
As a result, and can be respectively: 132,749,214,3200,14006 by summarizing the calculated result of each small task；It is finally completed
General assignment calculated result are as follows: 132+749+214+3200+14006=18301, as series digit (99,33,660,89,
106,108,200,3000,5006,9000) result of summation general assignment.
Specifically, host can for desktop computer, laptop, tablet computer, net book, super basis, smart phone,
One of server, but not limited to this；
Step (2): host acquisition can use slave list；Before the similar small task of fractionation is distributed to each slave by host, first
The slave of executable task is found, how host described in detail below obtains under complicated network environment can use slave list
Step method:
Step A1: host finds slave:
It includes being found by aggressive mode and/or being found by Passive Mode that host, which finds slave, and aggressive mode is dynamic for main owner
Slave is looked for, Passive Mode is that host passively waits slave to connect oneself.Aggressive mode can be scanning, indigo plant in local area network
One of tooth scanning, nfc card reading, remote Slave mode are a variety of；Passive Mode can be wide for LAN Access, bluetooth
It broadcasts, nfc card simulation, one of C/S model or a variety of.The specific method is as follows for various modes:
There are mainly two types of for the specific method of scan pattern in local area network:
I) host broadcasts UDP signal into local area network in port B1, and slave listening port B1 receives radio program, to realize master
The mutual discovery of machine and slave.
II) host calculates LAN IP range according to oneself IP and subnet mask, and then IP attempts connectivity port one by one
B1, slave listening port B1 wait connection, to realize that host and slave are mutually found.
The specific method of LAN Access mode is:
Host monitors some udp broadcast port B2, and slave is broadcasted to broadcast address port B2, then host can receive slave
Broadcast, to realize that host and slave are mutually found.
The specific method of C/S model is:
The IP that host has a slave that can directly be connected to, is set as hostIP, and host monitors some port B3(TCP/UDP), from
The port B3(TCP/UDP of machine connection hostIP), once successful connection, host and slave can mutually be found.
The specific method of bluetooth scan pattern is:
Host scans the slave for meeting certain feature, and Slave Broadcast meets the signal of host requirement feature, and host and slave pass through
Bluetooth realizes connecting communication.Premise using this kind of mode is that host and slave all support bluetooth.
The specific method of Bluetooth broadcast mode is:
Host broadcast has the signal of certain features, and slave scans the signal for meeting host broadcast feature, and host and slave use
Bluetooth connection communication.Premise using this kind of mode is that host and slave all support bluetooth.
The specific method of nfc card read mode is:
Host transmits data to slave, then executes writing operation as NFC reader, slave as simulation nfc card, host.It is main
Machine waits slave returned data, needs to read a secondary card at regular intervals according to the actual situation, to obtain latest data.
Premise using this kind of mode is that host and slave all support NFC.
The specific method of nfc card simulation model is:
Host is passive that NFC reader is waited to read data as simulation nfc card；Slave is NFC reader.Host is passed to slave
Transmission of data places the data in a data buffer queue, waits NFC reader to read, then deletes from buffer queue.
Host waits slave returned data, then the data of reader write-in are received as nfc card.
Premise using this kind of mode is that host and slave all support NFC.
The specific method of remote Slave mode is:
Host connects a preset server, obtains a slave IP+ port list；
Host uses IP and corresponding ports B4(TCP/UDP in list), slave monitors local port B4(TCP/UDP, and port needs
It is consistent with what is arranged in list), the connection from internet host is received, so that host be made to find slave.
Fig. 2 is the model selection flow chart that host of the present invention finds slave, as shown in Fig. 2, each aggressive mode or by dynamic model
The sequencing that formula enables is respectively as follows: scanning, LAN Access, C/S model, bluetooth scanning, Bluetooth broadcast, NFC in local area network
Card reading/nfc card simulation, remote Slave mode.The detailed step that each mode enables is as follows:
Step (211): scanning in local area network is successively opened first, udp broadcast port B2 is monitored, monitors the machine IP and port B3
(TCP/UDP), connection feedback result is waited, it is preferential to open and utilize this three kinds of searching modes, be conducive to improve connection speed；?
In the time of setting, if any connection feedback result, then the slave of connection is recorded into slave list；If the time beyond setting is still
Without as a result, then starting step (212) attempts bluetooth connection；
Step (212): first determining whether host itself supports Bluetooth function, if not supporting, is transferred to step (213) trial
NFC connection mode；If supporting, bluetooth scanning slave feature simultaneously attempts connection slave, within the set time, anti-if any connection
Feedback is as a result, be then recorded into slave list for the slave of connection, if the time beyond setting, it is blue to be still transferred to trial without feedback result
Tooth broadcast host feature simultaneously waits slave to connect, and within the set time, if any connection feedback result, then remembers the slave of connection
Typing slave list is equally transferred to step (213) and attempts NFC connection mode if time-out does not have connection；
Step (213): first determining whether host has NFC function, if not having, be transferred to step (214) attempt remotely from
Machine mode；If having, nfc card reading or nfc card simulation model are opened, and attempt to connect, it, then will connection if any successful connection
Slave be recorded into slave list, such as do not have successful connection, then be transferred to step (214) attempt remote Slave mode；
Step (214): host connects remote server, obtains slave list, if failed obtain, is transferred to step again
(211)；If successfully obtaining, be attached with the slave in slave list, if any the slave that can be connected to, then will can be connected to from
Machine is recorded into slave list, the slave that can be connected to if it does not exist, then is transferred to step (211) again.
More specifically, if host or slave can only use 2G/3G/4G network, it is required that: host must have public network IP,
Internet may have access to, and slave may have access to internet, and use C/S model.
More specifically, it if the connection network environment of host or slave is WiFi, the same local area network the case where, can be used
V2.4.1 or V2.4.2 version；The case where Different LANs, but slave can be connected to host, and V2.4.2 version can be used.
Step (22): host judges whether slave can be used:
It after host finds slave, further, also needs to judge the availability of slave, mainly be carried out by following two aspect
Judgement:
It 1) can connectivity judgement.If receiveed the response in time from function, it is unobstructed to show that host and slave can connect, slave has
Can connectivity, meet available condition.
2) whether slave has being currently running for task.It can be recorded by host oneself or slave actively illustrates that two ways is true
Determine whether slave has being currently running for task, if showing that slave meets available condition without being currently running for task.
Above two aspect can be all satisfied with condition, then slave is considered as available slave, be recorded as available slave, being included in can
Use slave list.
Step (3): N number of small task of fractionation is distributed to multiple available slaves and executes calculating by host；
Step (4): the task data that can be distributed with slave receiving host completes calculating task according to data processor, and anti-
Calculated result is presented to host；Specifically, slave can for desktop computer, laptop, tablet computer, net book, super basis,
One of smart phone, server, but not limited to this.More specifically, the mode that slave obtains data processor can be with
For one of following three kinds of modes:
A. host is distributed together with task data to slave, this kind of mode is relatively specific for small processing routine；
B. it is installed in advance on slave, this kind of mode is relatively specific for big common processing routine；
C. it is downloaded on host by slave, such mode is relatively specific for the processing routine of non-common big or middle type.
Three of the above acquisition modes can flexibly be selected according to different situations.
Specifically, host distribute small task to each slave and obtain each small task computation result detailed process it is as follows:
Host distributes one of them small task data (for example, (106,108) two number summations) to available slave list first
In one of slave, if slave refusal receive an assignment, show task send failure, then task data is redistributed to down
One slave；If slave receives an assignment, show that task is sent successfully, then carries out that inspection can be traversed with slave list；
If can not be traversed with slave list, continue to distribute next small task data (for example, (99,33) two number summations)
To the slave in available slave list；If can have been traversed with slave list, slave is waited to feed back calculated result；
When waiting slave feedback calculated result, if discovery has slave time-out not complete task, the company with the slave is disconnected in time
It connects, it is deleted from available slave list, while the task data that the slave executes is redistributed to other available slaves
It executes, slave list can be used by needing to obtain again at this time, then distributed tasks data；For the slave that time-out is rejected, in the case where carrying out
When distributed computing, if rediscovering it can be used, it is included in available slave list again；
When waiting slave feedback calculated result, when there is slave to complete before the deadline and feed back calculated result, meter is checked
It whether normal calculates result, if there is exception, the task data which executes is redistributed to other available slaves and is executed, this
When need to obtain again can be with slave list, then distributed tasks data；If result is normal, which is included in available slave and is arranged
Table；Later, it checks whether each small task is complete, if having unfinished, the unfinished small task data of distribution will be continued and arrived
Calculating can be executed with the slave in slave list, such as N number of small task has been fully completed calculating, then is transferred to step (5)；
Step (5): the calculated result that host is fed back according to each slave summarizes the calculating for completing general assignment.
Executed by the way that general assignment is split as multiple small tasks and distributes to multiple and different slaves, greatly reduce by
Single computer completes the consumption of the hardware and time resource of general assignment, while shortening the time for completing calculating task, improves
Task completes efficiency.
From the above technical scheme, a kind of distributed computing processing method of the invention, supports a plurality of types of calculating
Machine participates in setting up, and can support to work under the network environment of various complexity, can according to need real-time, conveniently, flexibly
Ground building, it is cheap, efficiently handle all kinds of large-scale tasks, so that all kinds of idle computer resources be enable to effectively integrate and fill
Divide and utilize, realizes the rational allocation and value maximization of resource.
The method of the present invention is easily achieved, adaptable, widely applicable using simple, helps to expand distributed computing
Suitable application area is conducive to the promotion and popularization for accelerating distributed computing technology.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, however the present invention is not limited thereto.All within the spirits and principles of the present invention, it is made it is any modification, equally replace
It changes and improves, should be included within the scope of the present invention.
Claims (9)
1. a kind of distributed computing processing method, which comprises the following steps:
Step (1): host receives general assignment, and general assignment is split into N number of small task, mutually indepedent between the small task；
Step (2): host acquisition can use slave list；
Step (3): N number of small task is distributed to multiple available slaves and executes calculating by host；
Step (4): the task data that can be distributed with slave receiving host completes calculating task according to data processor, and anti-
Calculated result is presented to host；
Step (5): the calculated result that host is fed back according to each slave summarizes the calculating for completing general assignment.
2. distributed computing processing method according to claim 1, which is characterized in that the host is desktop computer, pen
Remember one of this computer, tablet computer, net book, super basis, smart phone, server.
3. distributed computing processing method according to claim 1, which is characterized in that the slave is desktop computer, pen
Remember one of this computer, tablet computer, net book, super basis, smart phone, server.
4. distributed computing processing method according to claim 1, which is characterized in that the step (2) specifically includes:
Step (21): host finds slave；
Step (22): host judges whether slave can be used.
5. distributed computing processing method according to claim 4, which is characterized in that it includes logical that the host, which finds slave,
Aggressive mode is crossed to find and/or find by Passive Mode；The aggressive mode, which is that main owner is dynamic, looks for slave；It is described passive
Mode is that host passively waits slave to connect oneself.
6. distributed computing processing method according to claim 5, which is characterized in that the aggressive mode is in local area network
One of scanning, bluetooth scanning, nfc card reading, remote Slave mode are a variety of；The Passive Mode be LAN Access,
One of Bluetooth broadcast, nfc card simulation, C/S model are a variety of.
7. distributed computing processing method according to claim 6, which is characterized in that each aggressive mode or Passive Mode
The sequencing of enabling is respectively as follows: scanning, LAN Access, C/S model, bluetooth scanning, Bluetooth broadcast, nfc card in local area network
Reading/nfc card simulation, remote Slave mode.
8. distributed computing processing method according to claim 4, which is characterized in that the host judges that slave whether may be used
With specifically include it is following two aspect:
1) can connectivity judgement, i.e., whether host and slave can connect unobstructed；
2) whether slave has being currently running for task.
9. distributed computing processing method according to claim 1, which is characterized in that the source of the data processor
Mode is one of following three kinds of modes:
Host is distributed together with task data to slave；
It is installed in advance on slave；
It is downloaded on host by slave.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811468753.XA CN109597679A (en) | 2018-12-03 | 2018-12-03 | A kind of distributed computing processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811468753.XA CN109597679A (en) | 2018-12-03 | 2018-12-03 | A kind of distributed computing processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109597679A true CN109597679A (en) | 2019-04-09 |
Family
ID=65960718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811468753.XA Pending CN109597679A (en) | 2018-12-03 | 2018-12-03 | A kind of distributed computing processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109597679A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132929A (en) * | 2021-03-15 | 2021-07-16 | 深圳市晨北科技有限公司 | Equipment network access method, device, computer equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105471950A (en) * | 2014-09-05 | 2016-04-06 | 鸿富锦精密工业(武汉)有限公司 | Distributed computing method and system |
| CN106598735A (en) * | 2016-12-13 | 2017-04-26 | 广东金赋科技股份有限公司 | Distributive calculation method, main control node, calculation node and system |
| CN106648900A (en) * | 2016-12-28 | 2017-05-10 | 深圳Tcl数字技术有限公司 | Smart television-based supercomputing method and system |
-
2018
- 2018-12-03 CN CN201811468753.XA patent/CN109597679A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105471950A (en) * | 2014-09-05 | 2016-04-06 | 鸿富锦精密工业(武汉)有限公司 | Distributed computing method and system |
| CN106598735A (en) * | 2016-12-13 | 2017-04-26 | 广东金赋科技股份有限公司 | Distributive calculation method, main control node, calculation node and system |
| CN106648900A (en) * | 2016-12-28 | 2017-05-10 | 深圳Tcl数字技术有限公司 | Smart television-based supercomputing method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132929A (en) * | 2021-03-15 | 2021-07-16 | 深圳市晨北科技有限公司 | Equipment network access method, device, computer equipment and storage medium |
| CN113132929B (en) * | 2021-03-15 | 2023-10-27 | 深圳市晨北科技有限公司 | Equipment network access method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103414761B (en) | Mobile terminal cloud resource scheduling method based on Hadoop framework | |
| US10511482B2 (en) | Dynamic activation of web applications | |
| CN106850788B (en) | Integrated framework and integrated approach towards multi-source heterogeneous geographic information resources | |
| CN101436148B (en) | Integrated client end and method for performing interaction of desktop application and network WEB application | |
| CN104391918B (en) | The implementation method of distributed networks database query priority management based on equity deployment | |
| CN104506620A (en) | Extensible automatic computing service platform and construction method for same | |
| CN106686132A (en) | Yunhai system deployment method and system | |
| KR20210036226A (en) | A distributed computing system including multiple edges and cloud, and method for providing model for using adaptive intelligence thereof | |
| CN106033476A (en) | Incremental graphic computing method in distributed computing mode under cloud computing environment | |
| CN106412009A (en) | Interface calling method and device | |
| CN103399894A (en) | Distributed transaction processing method on basis of shared storage pool | |
| CN108989385A (en) | A kind of implementation method based on Zabbix monitoring collection automatic synchronization CMDB | |
| CN114666335B (en) | Distributed system load balancing device based on data distribution service DDS | |
| CN101771724B (en) | Heterogeneous distributed information integration method, device and system | |
| CN111240822B (en) | Task scheduling method, device, system and storage medium | |
| CN112351106B (en) | Service grid platform containing event grid and communication method thereof | |
| CN109597679A (en) | A kind of distributed computing processing method | |
| CN103338160B (en) | Based on optimization system and the optimization method of the polling service of buffer queue | |
| CN105892957B (en) | A kind of distributed transaction execution method based on Dynamic Program Slicing | |
| CN104166581A (en) | Virtualization method for increment manufacturing device | |
| CN103152428A (en) | Method for carrying out service communication among nodes on cloud platform | |
| CN106447755A (en) | Animation rendering system | |
| CN102882709A (en) | Monitoring system and monitoring method in cloud computing environment | |
| CN109669788A (en) | The MPI implementation method of multi core chip towards direct memory access interconnection communication | |
| CN109343967A (en) | A kind of distributed computing system and calculation method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190409 |