CN116662283A - Data sharing and calculating method with end-edge cooperation - Google Patents
Data sharing and calculating method with end-edge cooperation Download PDFInfo
- Publication number
- CN116662283A CN116662283A CN202310937222.5A CN202310937222A CN116662283A CN 116662283 A CN116662283 A CN 116662283A CN 202310937222 A CN202310937222 A CN 202310937222A CN 116662283 A CN116662283 A CN 116662283A
- Authority
- CN
- China
- Prior art keywords
- data
- side device
- calculation
- mode
- side equipment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 90
- 238000004364 calculation method Methods 0.000 claims abstract description 120
- 230000008569 process Effects 0.000 claims abstract description 58
- 230000005540 biological transmission Effects 0.000 claims abstract description 29
- 238000005520 cutting process Methods 0.000 claims abstract description 7
- 238000004891 communication Methods 0.000 claims description 20
- 230000011218 segmentation Effects 0.000 claims description 19
- 238000012795 verification Methods 0.000 claims description 9
- 238000013524 data verification Methods 0.000 abstract description 5
- 230000006870 function Effects 0.000 abstract description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/17—Details of further file system functions
- G06F16/176—Support for shared access to files; File sharing support
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention belongs to the technical field of edge calculation, and particularly discloses a data sharing and calculating method for end-edge cooperation. According to different distributed computing scenes, the method designs a capacity cooperative mode and a format cooperative mode, and meets the requirements of data sharing and data computing in different distributed computing scenes. In addition, the method designs a calculation mode of calculating force when different instruction set devices coexist and a mode of automatically cutting data under different calculating force under the capacity cooperative mode, and achieves the functions of self-adaptive cutting and transmission of the data according to the calculating force of the different instruction set devices in a distributed calculation scene. In addition, the method designs different device access rights, data transmission directions and data verification processes in the capacity cooperative mode and the format cooperative mode, and ensures the data transmission efficiency and the data safety on the premise of meeting the data sharing and transmission requirements in the different modes.
Description
Technical Field
The invention belongs to the technical field of edge calculation, and particularly relates to a data sharing and calculating method for end-edge cooperation.
Background
Along with the rapid development of the front-edge technologies such as the Internet of things, the 5G network and the artificial intelligence, the capability of processing data on the terminal side equipment is greatly improved, so that a mode of combining terminal side calculation with edge calculation is adopted in more and more application scenes, a mixed mode of the edge and the terminal side is formed, more efficient data processing and distribution are realized, and different calculation demands are met.
However, in the traditional end-side collaborative computing framework, the computing processes of the end side and the side are either separated or serial computing processes, the end side device is more burdened with small computing amount or data preprocessing work, and the large-scale data computing is more burdened with the side or cloud side. With the development of the chip towards the directions of miniaturization, high frequency and high computation, more and more distributed computation scenes of parallel computation of the end side and the side occur in the future end-side collaboration, but the end-side collaboration framework of the current period obviously cannot meet the requirements of data sharing and data computation in some distributed computation scenes.
Disclosure of Invention
In order to solve at least one of the problems mentioned in the background art, the present invention proposes a data sharing and computing method with end-to-end coordination.
A data sharing and calculating method of end-edge cooperation comprises the following steps:
step S1, constructing a communication protocol and an operation mode of an edge device and an end device, and writing the constructed operation mode and the communication protocol into the edge device and the end device, specifically, the communication protocol includes: designating access rights and write rights of the side device and the end device;
the working mode comprises a capacity cooperative mode and a format cooperative mode, wherein in the capacity cooperative mode, the end side equipment and the side equipment have access rights and writing rights; in the format collaboration mode, the side device has access rights to the end device and the end device, and has no write rights.
Step S2, the side device and the end device establish a communication connection according to the set operation mode, specifically,
if the working mode is a capacity cooperative mode, establishing a full duplex communication mode by the end side equipment and the side equipment, wherein the data transmission mode is parallel transmission;
if the working mode is a format cooperative mode, a half-duplex communication mode is established by the end side equipment and the side equipment, and the data transmission mode is parallel transmission;
step S3, the terminal side equipment collects data and executes different data sharing modes through the set different working modes, and the method specifically comprises the steps of:
step S301, the terminal device collects data, and stores the collected data in a cache;
step S302, according to the set different operation modes, different data sharing processes are executed respectively, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
step A1, the end side device sends the hardware parameters to the side device, and the side device calculates the calculation force of different devices according to the hardware parameters and the hardware parameters of the end side device, wherein the calculation formula of the specific calculation force is as follows:
in the method, in the process of the invention,representing the computational power of the device, +.>Representing the number of the computing unit in the device, +.>Representation deviceNumber of->Representing the computational power of the corresponding computational unit, +.>Weight value representing calculation unit is experience number, < ->Representing the number of computational cores in the computational unit, +.>Representing the number of operations of the computational core in a cycle, < >>Indicating the operating frequency of the computing unit.
Step A2, the side equipment performs data segmentation calculation according to different calculation forces, and sends the calculated end side equipment data segmentation size to the end side equipment, wherein a specific data segmentation calculation formula is as follows:
in the method, in the process of the invention,indicating the magnitude of the calculation force of the side device, < >>Representing the calculated force of the end side device, M representing the total number of side devices and end side devices, +.>Indicates the total number of sizes, +.>Data size representing side device division, +.>The data size of the terminal equipment is represented; />Indicating the number of the side device,indicating the number of the end-side device.
Step A3, the terminal side equipment receives the data segmentation size sent by the side equipment, segments the data in the memory according to the value of the data segmentation size, and packages the segmented data into structural information;
step A4, the end side equipment transmits the segmented data and the data structuring information I to the side equipment;
and step A5, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one round of verification on the data according to the analyzed content, wherein the specific data verification flow is as follows:
firstly, analyzing the data structure information, and analyzing the data format and the data size of received data from the data structure information;
then scanning the storage position by the side, and reading the data format and the data size of the stored data;
and finally, comparing the data format and the data size of the read storage data, and analyzing whether the data format and the data size of the received data are consistent.
If the working mode is the format cooperative mode, executing the steps of:
step B1, the side device sends the calculated data format to the end device, the end device reads the format of the data in the cache, and segments the data according to the data format, specifically,
the executing process in the cutting process is to read the data in the memory through the terminal side equipment and judge the format of the data, and then load the data in the specific format in the memory according to the preset data format of the terminal side equipment;
splitting the split data, namely writing the data into a computing unit and executing a computing process for the data needing to be loaded by the side equipment, and continuously storing the data needing to be loaded by the side equipment in a memory for the next data transmission process.
Step B2, the end side device packages the segmented data into structural information, and sends the segmented data and the data structural information I to the side device;
and B3, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one round of verification on the data according to the analyzed content, wherein the specific process comprises the following steps of:
firstly, analyzing the data structure information, and analyzing the data format and the data size of received data from the data structure information;
then the side equipment scans the received data and reads the data format and the data size of the stored data;
and finally judging whether errors occur in the data transmission and storage processes by judging whether the read data format and the data size are consistent with the data format and the data size of the parsed data.
Step S4, the end device and the side device calculate the shared data according to different working modes, and specifically includes the steps of:
step S401, the end device and the side device read data according to the instruction, and load the data into the memory;
step S402, according to the set different operation modes, respectively executing different data calculation flows, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
step C1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step C2, the side device updates its own calculation process by accessing the parameters in the side device memory, and correspondingly, the side device updates its own calculation process by accessing the parameters in the side device memory;
the specific updating mode is as follows: the parameters in the side equipment and the end equipment are calculated by updating the opposite side equipment through non-logic operation with the data in the opposite side equipment, the parameters participate in the subsequent calculation steps, and the final numerical value is written into the opposite side equipment;
step C3, the end side device and the side device execute the calculation process to obtain a calculation result, the end side device transmits the calculation result to the side device, the side device combines the calculation result transmitted by the end side device with the calculation result of the side device, and the combined calculation result is stored to the local;
if the working mode is the format cooperative mode, executing the steps of:
step D1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step D2, the end device updates its own calculation process by accessing the parameters in the memory of the side device, and correspondingly, the side device updates its own calculation process by accessing the parameters in the memory of the end device;
the specific updating mode is as follows: the parameters in the side equipment and the end side equipment are calculated by carrying out logic operation with the data in the opposite side equipment to update the calculation process of the opposite side equipment, the parameters do not participate in the subsequent calculation steps, and the final numerical value is not written into the opposite side equipment;
and D3, after the end side device and the side device execute the calculation process, obtaining a calculation result, transmitting the calculation result to the side device by the end side device, combining the calculation result of the end side device with the calculation result of the side device by the side device, and storing the combined calculation result to the local.
The invention provides a data sharing and calculating method with end-edge cooperation, which has the following beneficial effects compared with the prior art:
the invention provides a method for sharing and calculating data under the coordination of end edges, which designs a capacity coordination mode and a format coordination mode according to different distributed calculation scenes, and meets the requirements of data sharing and data calculation under different distributed calculation scenes;
according to the method provided by the invention, a calculation mode of calculating force when different instruction set devices coexist in a capacity cooperative mode is designed, and a mode of automatically segmenting data according to different calculation force sizes by side devices and end devices is provided, so that the functions of adaptively segmenting and transmitting data according to the calculation force sizes of different devices in a distributed calculation scene when different calculation force devices exist are realized;
the method provided by the invention designs different equipment access rights, data transmission directions and data verification processes in the capacity cooperative mode and the format cooperative mode, and ensures the data transmission efficiency and the data safety on the premise of meeting the data sharing and transmission requirements in different modes.
Drawings
Fig. 1 is a system flow diagram of the present invention.
Detailed Description
In order to make the objects and features of the present invention more comprehensible, the present invention is described in detail below by way of examples and with reference to the accompanying drawings.
As shown in fig. 1, a method for sharing and calculating data by end-edge cooperation includes the following steps:
step S1, constructing a communication protocol and an operation mode of an edge device and an end device, and writing the constructed operation mode and the communication protocol into the edge device and the end device, specifically, the communication protocol includes: and designating the access rights and the write rights of the side device and the end device.
The working mode comprises a capacity cooperative mode and a format cooperative mode, wherein in the capacity cooperative mode, the end side equipment and the side equipment have access rights and writing rights; in the format collaboration mode, the side device has access rights to the end device and the end device, and has no write rights.
Aiming at the problem that the requirements of data sharing and data computing in some distributed computing scenes can not be met under the traditional end-edge collaboration framework, the invention provides a method for data sharing and computing under the end-edge collaboration.
Optionally, in the capacity cooperative mode, the data types exchanged by the end-side device and the side-side device are the same; the data types exchanged by the end-side device and the side-side device in the format cooperative mode are not necessarily the same.
Optionally, the operation mode includes, in addition to the capacity cooperative mode and the format cooperative mode, an independent serial mode in which the side device has access rights and write rights to the end-side device and the end-side device does not have access rights and write rights to the side device.
Step S2, the side device and the end device establish a communication connection according to the set operation mode, specifically,
if the working mode is a capacity cooperative mode, establishing a full duplex communication mode by the end side equipment and the side equipment, wherein the data transmission mode is parallel transmission;
if the working mode is a format cooperative mode, a half-duplex communication mode is established by the end side equipment and the side equipment, and the data transmission mode is parallel transmission;
aiming at different emphasis of data sharing and data calculation in different collaborative modes, the invention designs different equipment access rights, data transmission directions and data verification processes in the capacity collaborative mode and the format collaborative mode, and ensures the data transmission efficiency and the data safety on the premise of meeting the data sharing and transmission requirements in different modes.
Optionally, the side device and the end device establish communication connection, and the number of the end devices can be configured in a section according to the calculation force of the side device, and can be estimated according to the actual calculation force or configured according to an empirical value.
Step S3, the terminal side equipment collects data and executes different data sharing modes through the set different working modes, and the method specifically comprises the steps of:
in step S301, the end-side device collects data, and stores the collected data in a cache.
Step S302, according to the set different operation modes, different data sharing processes are executed respectively, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
step A1, the end side device sends the hardware parameters to the side device, and the side device calculates the calculation force of different devices according to the hardware parameters and the hardware parameters of the end side device, wherein the calculation formula of the specific calculation force is as follows:
in the method, in the process of the invention,representing the computational power of the device, +.>Representing the number of the computing unit in the device, +.>Indicating the number of the device>Representing the computational power of the corresponding computational unit, +.>Weight value representing calculation unit is experience number, < ->Representation ofThe number of computation cores in the computation unit, +.>Representing the number of operations of the computational core in a cycle, < >>Indicating the operating frequency of the computing unit.
The traditional edge collaborative framework has no data segmentation requirement, and for some non-edge collaborative distributed computing scenes, data is distributed according to the load of threads.
Step A2, the side equipment performs data segmentation calculation according to different calculation forces, and sends the calculated end side equipment data segmentation size to the end side equipment, wherein a specific data segmentation calculation formula is as follows:
in the method, in the process of the invention,indicating the magnitude of the calculation force of the side device, < >>Representing the calculated force of the end side device, M representing the total number of side devices and end side devices, +.>Indicates the total number of sizes, +.>Data size representing side device division, +.>The data size of the terminal equipment is represented; />Indicating the number of the side device,indicating the number of the end-side device.
On the basis of providing calculation modes of calculation forces of different instruction set devices, the invention designs a mode of automatically cutting data according to the different calculation forces by the side device and the end device, and realizes the functions of self-adaptive cutting and transmission of data according to the calculation forces of different devices in a distributed calculation scene when the different calculation forces exist.
And step A3, the terminal side equipment receives the data segmentation size sent by the side equipment, segments the data in the memory according to the value of the data segmentation size, and packages the segmented data into structural information.
Optionally, the structured information adopts json format, and the specific information is:
info= { "data position", "data format", "data size", "others" }
Step A4, the end side equipment transmits the segmented data and the data structuring information I to the side equipment;
and step A5, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one round of verification on the data according to the analyzed content, wherein the specific process comprises the following steps:
firstly, analyzing the data structure information, and analyzing the data format and the data size of received data from the data structure information;
then the side equipment scans the received data and reads the data format and the data size of the stored data;
and finally judging whether errors occur in the data transmission and storage processes by judging whether the read data format and the data size are consistent with the data format and the data size of the parsed data.
Generally, in a distributed computing scenario, because the data sharing among different devices lacks a necessary verification mechanism, if errors exist in a data segmentation stage, the data computing errors caused by the lack of effective supervision when the different devices share the data cannot be perceived and positioned.
If the working mode is the format cooperative mode, executing the steps of:
step B1, the side device sends the calculated data format to the end device, the end device reads the format of the data in the cache, and segments the data according to the data format, specifically,
the executing process in the cutting process is to read the data in the memory through the terminal side equipment and judge the format of the data, and then load the data in the specific format in the memory according to the preset data format of the terminal side equipment.
Splitting the split data, namely writing the data into a computing unit and executing a computing process for the data needing to be loaded by the side equipment, and continuously storing the data needing to be loaded by the side equipment in a memory for the next data transmission process.
Step B2, the end side device packages the segmented data into structural information, and sends the segmented data and the data structural information I to the side device;
optionally, the structured information adopts json format, and the specific information is:
info= { "data position", "data format", "data size", "others" }
And B3, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one round of verification on the data according to the analyzed content, wherein the specific data verification flow is as follows:
firstly, analyzing the data structure information, and analyzing the data format and the data size of the received data from the data structure information.
And then scanning the storage position by the side, and reading the data format and the data size of the stored data.
And finally, comparing the data format and the data size of the read storage data, and analyzing whether the data format and the data size of the received data are consistent.
Similarly, in a distributed computing scenario, because the data sharing among different devices lacks a necessary verification mechanism, if errors exist in a data segmentation stage, the data computing errors caused by the lack of effective supervision when the different devices share the data cannot be perceived and positioned.
Optionally, if the working mode is an independent serial mode, the terminal side device performs preprocessing on the data first, then sends the preprocessed data to the side device, and the side device completes the calculation process of the data.
Step S4, the end device and the side device calculate the shared data according to different working modes, and specifically includes the steps of:
step S401, the end device and the side device read data according to the instruction, and load the data into the memory;
step S402, according to the set different operation modes, respectively executing different data calculation flows, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
step C1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step C2, the side device updates its own calculation process by accessing the parameters in the side device memory, and correspondingly, the side device updates its own calculation process by accessing the parameters in the side device memory;
the specific updating mode is as follows: the parameters in the side equipment and the end equipment are calculated by updating the opposite side equipment through non-logic operation with the data in the opposite side equipment, the parameters participate in the subsequent calculation steps, and the final numerical value is written into the opposite side equipment;
step C3, the end side device and the side device execute the calculation process to obtain a calculation result, the end side device transmits the calculation result to the side device, the side device combines the calculation result transmitted by the end side device with the calculation result of the side device, and the combined calculation result is stored to the local;
if the working mode is the format cooperative mode, executing the steps of:
step D1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step D2, the end device updates its own calculation process by accessing the parameters in the memory of the side device, and correspondingly, the side device updates its own calculation process by accessing the parameters in the memory of the end device;
the specific updating mode is as follows: the parameters in the side equipment and the end side equipment are calculated by carrying out logic operation with the data in the opposite side equipment to update the calculation process of the opposite side equipment, the parameters do not participate in the subsequent calculation steps, and the final numerical value is not written into the opposite side equipment;
and D3, after the end side device and the side device execute the calculation process, obtaining a calculation result, transmitting the calculation result to the side device by the end side device, combining the calculation result of the end side device with the calculation result of the side device by the side device, and storing the combined calculation result to the local.
While the invention has been described in detail in this specification with reference to the general description and the specific embodiments thereof, it will be apparent to one skilled in the art that modifications and improvements can be made thereto. Accordingly, such modifications or improvements may be made without departing from the spirit of the invention and are intended to be within the scope of the invention as claimed.
Claims (10)
1. The end-edge cooperative data sharing and calculating method is characterized by comprising the following steps of:
step S1, constructing communication protocols and working modes of side equipment and end equipment, and writing the constructed working modes and communication protocols into the side equipment and the end equipment;
step S2, the side equipment and the end side equipment establish communication connection according to a set working mode;
step S3, the terminal side equipment collects data and executes different data sharing processes according to different working modes;
and S4, the end side device and the side device execute different data calculation flows on the shared data according to different working modes.
2. The method for sharing and calculating data by end-edge coordination according to claim 1, wherein the constructing a communication protocol between the end-edge device and the end-edge device in step S1 includes: designating access rights and write rights of the side device and the end device;
the working mode comprises a capacity cooperative mode and a format cooperative mode, wherein in the capacity cooperative mode, the end side equipment and the side equipment have access rights and writing rights; in the format collaboration mode, the side device has access rights to the end device and the end device, and has no write rights.
3. The method for sharing and calculating data with end-side coordination according to claim 1, wherein said side device and said end-side device in step S2 establish a communication connection according to a set operation mode, wherein,
if the working mode is a capacity cooperative mode, establishing a full duplex communication mode by the end side equipment and the side equipment, wherein the data transmission mode is parallel transmission;
if the working mode is a format cooperative mode, a half-duplex communication mode is established by the end side device and the side device, and the data transmission mode is parallel transmission.
4. The method for sharing and calculating data by end-edge coordination according to claim 1, wherein step S3 comprises:
step S301, the terminal device collects data, and stores the collected data in a cache;
step S302, according to the set different operation modes, different data sharing processes are executed respectively, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
a1, the end side equipment sends the hardware parameters to the side equipment, and the side equipment calculates the calculation force of different equipment according to the hardware parameters and the hardware parameters of the end side equipment;
step A2, the side equipment performs data segmentation calculation according to different calculation forces, and sends the calculated end side equipment data segmentation size to the end side equipment;
step A3, the terminal side equipment receives the data segmentation size sent by the side equipment, segments the data in the memory according to the value of the data segmentation size, and packages the segmented data into structural information;
step A4, the end side equipment transmits the segmented data and the data structuring information I to the side equipment;
step A5, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one round of verification on the data according to the analyzed content;
if the working mode is the format cooperative mode, executing the steps of:
step B1, the side equipment sends the calculated data format to the end side equipment, the end side equipment reads the format of the data in the cache and segments the data according to the format of the data;
step B2, the end side device packages the segmented data into structural information, and sends the segmented data and the data structural information I to the side device;
and B3, the side equipment receives the data and the data structure information sent by the end side equipment, analyzes the data structure information, and performs one-round verification on the data according to the analyzed content.
5. The method for sharing and calculating data by end-to-end coordination according to claim 4, wherein in the step A1, the computing force of different devices is calculated by the side device according to the hardware parameters of the side device and the hardware parameters of the end-to-end device, and a specific computing formula of the computing force is as follows:
；
in the method, in the process of the invention,representing the computational power of the device, +.>Representing the number of the computing unit in the device, +.>Indicating the number of the device>Representing the computational power of the corresponding computational unit, +.>Weight value representing calculation unit is experience number, < ->Representing the number of computational cores in the computational unit, +.>Representing the number of operations of the computational core in a cycle, < >>Indicating the operating frequency of the computing unit.
6. The method for sharing and calculating data by end-to-end coordination according to claim 4, wherein the data segmentation calculation is performed by the side device in step A2 according to different calculation forces, and a specific data segmentation calculation formula is as follows:
；
in the method, in the process of the invention,indicating the magnitude of the calculation force of the side device, < >>Representing the calculated force of the end side device, M representing the total number of side devices and end side devices, +.>Indicates the total number of sizes, +.>Data size representing side device division, +.>The data size of the terminal equipment is represented; />Number indicating side device, +.>Indicating the number of the end-side device.
7. The method for sharing and computing data with end-to-end coordination according to claim 4, wherein the data is sliced according to the format of the data in step B1, wherein,
the executing process in the cutting process is to read the data in the memory through the terminal side equipment and judge the format of the data, and then load the data in the specific format in the memory according to the preset data format of the terminal side equipment;
and splitting the split data, namely writing the data into a computing unit and executing a computing process when the split data is the data needing to be loaded by the side equipment, and continuously storing the split data into a memory to wait for the next data transmission process when the split data is the data needing to be loaded by the side equipment.
8. The method for sharing and calculating data by end-to-end coordination according to claim 4, wherein the step A5 and the step B3 perform a round of verification on the data, and the specific process includes:
firstly, analyzing the data structure information, and analyzing the data format and the data size of received data from the data structure information;
then the side equipment scans the received data and reads the data format and the data size of the stored data;
and finally judging whether errors occur in the data transmission and storage processes by judging whether the read data format and the data size are consistent with the parsed data format and the data size.
9. The method for sharing and calculating data by end-edge coordination according to claim 1, wherein the step S4 comprises the steps of:
step S401, the end device and the side device read data according to the instruction, and load the data into the memory;
step S402, according to the set different operation modes, respectively executing different data calculation flows, wherein,
if the working mode is the capacity cooperative mode, executing the steps of:
step C1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step C2, the side device updates its own calculation process by accessing the parameters in the side device memory, and correspondingly, the side device updates its own calculation process by accessing the parameters in the side device memory;
step C3, the end side device and the side device execute the calculation process to obtain a calculation result, the end side device transmits the calculation result to the side device, the side device combines the calculation result transmitted by the end side device with the calculation result of the side device, and the combined calculation result is stored to the local;
if the working mode is the format cooperative mode, executing the steps of:
step D1, the end side device and the side device read the data in the memory and calculate, and store the parameters in the calculation process in the respective caches;
step D2, the end device updates its own calculation process by accessing the parameters in the memory of the side device, and correspondingly, the side device updates its own calculation process by accessing the parameters in the memory of the end device;
and D3, after the end side device and the side device execute the calculation process, obtaining a calculation result, transmitting the calculation result to the side device by the end side device, combining the calculation result of the end side device with the calculation result of the side device by the side device, and storing the combined calculation result to the local.
10. The method for data sharing and computing in cooperation with end-edge of claim 9, wherein the update itself computing in step C2 is different from the update itself computing in step D2, wherein,
in the calculation process of updating itself in step C2, parameters in the side device and the end device are calculated by performing non-logical operation with data in the opposite device to update the calculation process of the opposite device, the parameters participate in the subsequent calculation steps, and the final values are written into the opposite device;
in the calculation process of updating itself described in step D2, the parameters in the side device and the end device are calculated by performing a logical operation with the data in the opposite device to update the calculation process of the opposite device, the parameters do not participate in the subsequent calculation steps, and the final values are not written into the opposite device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310937222.5A CN116662283B (en) | 2023-07-28 | 2023-07-28 | Data sharing and calculating method with end-edge cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310937222.5A CN116662283B (en) | 2023-07-28 | 2023-07-28 | Data sharing and calculating method with end-edge cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116662283A true CN116662283A (en) | 2023-08-29 |
| CN116662283B CN116662283B (en) | 2023-12-12 |
Family
ID=87720929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310937222.5A Active CN116662283B (en) | 2023-07-28 | 2023-07-28 | Data sharing and calculating method with end-edge cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116662283B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018121282A1 (en) * | 2016-12-26 | 2018-07-05 | 华为技术有限公司 | Data processing method, end device, cloud device, and end-cloud collaboration system |
| CN110851255A (en) * | 2019-11-07 | 2020-02-28 | 之江实验室 | Method for processing video stream based on cooperation of terminal equipment and edge server |
| CN112287609A (en) * | 2020-12-28 | 2021-01-29 | 之江实验室 | End, edge and cloud collaborative computing device for robot task division |
| DE102020208023A1 (en) * | 2019-09-28 | 2021-04-01 | Intel Corporation | ADAPTIVE DATA FLOW TRANSFORMATION IN EDGE COMPUTING ENVIRONMENTS |
| CN112926660A (en) * | 2021-02-26 | 2021-06-08 | 武汉大学 | Water level identification system and method with cooperative end edges |
| CN114327858A (en) * | 2021-11-12 | 2022-04-12 | 中国科学院计算技术研究所 | Cloud edge end distributed computing power cooperation method and system based on control domain |
| CN114579318A (en) * | 2022-05-06 | 2022-06-03 | 北京智芯微电子科技有限公司 | Resource coordination method, device and equipment for edge computing |
| KR20220078411A (en) * | 2020-12-03 | 2022-06-10 | 한국전자기술연구원 | Edge computing node and method for sharing data thereof |
| CN114661466A (en) * | 2022-03-21 | 2022-06-24 | 东南大学 | Task unloading method for intelligent workflow application in edge computing environment |
| CN115080210A (en) * | 2022-06-29 | 2022-09-20 | 重庆长安汽车股份有限公司 | Vehicle-mounted computing power sharing system, method, device and medium based on edge computing |
| KR20220168527A (en) * | 2021-06-15 | 2022-12-23 | 연세대학교 산학협력단 | Apparatus and method for preserving privacy in edge-server synergetic computing |
| CN116436980A (en) * | 2022-12-20 | 2023-07-14 | 之江实验室 | Real-time video task end network edge cooperative scheduling method and device |
-
2023
- 2023-07-28 CN CN202310937222.5A patent/CN116662283B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018121282A1 (en) * | 2016-12-26 | 2018-07-05 | 华为技术有限公司 | Data processing method, end device, cloud device, and end-cloud collaboration system |
| DE102020208023A1 (en) * | 2019-09-28 | 2021-04-01 | Intel Corporation | ADAPTIVE DATA FLOW TRANSFORMATION IN EDGE COMPUTING ENVIRONMENTS |
| CN110851255A (en) * | 2019-11-07 | 2020-02-28 | 之江实验室 | Method for processing video stream based on cooperation of terminal equipment and edge server |
| KR20220078411A (en) * | 2020-12-03 | 2022-06-10 | 한국전자기술연구원 | Edge computing node and method for sharing data thereof |
| CN112287609A (en) * | 2020-12-28 | 2021-01-29 | 之江实验室 | End, edge and cloud collaborative computing device for robot task division |
| CN112926660A (en) * | 2021-02-26 | 2021-06-08 | 武汉大学 | Water level identification system and method with cooperative end edges |
| KR20220168527A (en) * | 2021-06-15 | 2022-12-23 | 연세대학교 산학협력단 | Apparatus and method for preserving privacy in edge-server synergetic computing |
| CN114327858A (en) * | 2021-11-12 | 2022-04-12 | 中国科学院计算技术研究所 | Cloud edge end distributed computing power cooperation method and system based on control domain |
| CN114661466A (en) * | 2022-03-21 | 2022-06-24 | 东南大学 | Task unloading method for intelligent workflow application in edge computing environment |
| CN114579318A (en) * | 2022-05-06 | 2022-06-03 | 北京智芯微电子科技有限公司 | Resource coordination method, device and equipment for edge computing |
| CN115080210A (en) * | 2022-06-29 | 2022-09-20 | 重庆长安汽车股份有限公司 | Vehicle-mounted computing power sharing system, method, device and medium based on edge computing |
| CN116436980A (en) * | 2022-12-20 | 2023-07-14 | 之江实验室 | Real-time video task end network edge cooperative scheduling method and device |
Non-Patent Citations (2)
| Title |
|---|
| 于清林;: "从边缘计算到算力网络", 产业科技创新, no. 03 * |
| 邓晓衡;关培源;万志文;刘恩陆;罗杰;赵智慧;刘亚军;张洪刚;: "基于综合信任的边缘计算资源协同研究", 计算机研究与发展, no. 03 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116662283B (en) | 2023-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112181971B (en) | Edge-based federated learning model cleaning and equipment clustering method and system | |
| US20210042578A1 (en) | Feature engineering orchestration method and apparatus | |
| CN215378952U (en) | Edge gateway terminal equipment and data processing system based on cloud edge cooperation technology | |
| CN104461551A (en) | Parallel data processing based big data processing system | |
| Zhang et al. | Risk-aware cloud-edge computing framework for delay-sensitive industrial IoTs | |
| CN112241331A (en) | Mobile Web task dynamic migration method based on deep neural network | |
| CN116662283B (en) | Data sharing and calculating method with end-edge cooperation | |
| CN114584429A (en) | Industrial intelligent internet of things gateway | |
| CN112235387B (en) | Multi-node cooperative computing unloading method based on energy consumption minimization | |
| CN108762068A (en) | A kind of multiple agent consistency control method with model uncertainty | |
| CN112835691A (en) | Edge data processing method and system for communication of Internet of things | |
| Han et al. | Research on multinode collaborative computing offloading algorithm based on minimization of energy consumption | |
| WO2019127926A1 (en) | Calculation method and calculation device for sparse neural network, electronic device, computer readable storage medium, and computer program product | |
| CN112787835B (en) | Network device and method for processing data related to network message | |
| EP4311202A1 (en) | End-edge-cloud coordination system and method based on digital retina, and device | |
| CN113419842A (en) | Method and device for constructing edge computing microservice based on JavaScript | |
| CN209803775U (en) | Data processing apparatus | |
| CN113473515A (en) | Industrial wireless network reliability assessment method | |
| CN112333425A (en) | Communication system for ship networking | |
| CN214228277U (en) | Electromechanical device monitoring system and Internet of things host | |
| CN115087042A (en) | 5G private network data distribution method, device, equipment and storage medium | |
| CN115473861B (en) | High-performance processing system and method based on communication and calculation separation and storage medium | |
| US11887353B1 (en) | Deep learning image classification oriented to heterogeneous computing device | |
| CN212572621U (en) | Cloud valley data terminal | |
| CN116737347B (en) | Task scheduling control 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |