CN110784512A - Airborne dynamic cloud system and real-time response resource allocation method thereof - Google Patents
Airborne dynamic cloud system and real-time response resource allocation method thereof Download PDFInfo
- Publication number
- CN110784512A CN110784512A CN201910878010.8A CN201910878010A CN110784512A CN 110784512 A CN110784512 A CN 110784512A CN 201910878010 A CN201910878010 A CN 201910878010A CN 110784512 A CN110784512 A CN 110784512A
- Authority
- CN
- China
- Prior art keywords
- robot
- resource
- real
- robots
- time
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
Abstract
The invention relates to an airborne dynamic cloud system and a real-time response resource allocation method thereof, wherein the airborne dynamic cloud integrates airborne idle computing resources of a plurality of robots to form a dynamic and real-time cloud system; the real-time response resource management method comprises a computational utility bandwidth allocation method and a real-time response task allocation method, and provides real-time cloud service support for robots in a group robot system when executing computation-intensive tasks. Compared with the prior art, the invention aims to enable the group of robots to share computing resources as required and provide a high-performance airborne dynamic cloud for computing-intensive tasks without depending on a cloud server and communication infrastructure.
Description
Technical Field
The invention relates to a real-time cloud service support in the field of group robots, in particular to an airborne dynamic cloud system and a real-time response resource allocation method thereof.
Background
A single robot tends to be difficult to independently and well perform tasks in complex environments: when a task is executed, once a fault occurs, the task return can be terminated immediately, and an immeasurable loss is likely to be caused. With the rapid development of computer miniaturization and communication technology, the use of group robots is beneficial to improving the accuracy, robustness and the like of task execution in a complex environment.
The computing resources needed by the robot when facing different tasks often have great differences, for example, the data processing pressure faced by the robot in the tasks of cluster motion, navigation and the like is small, and the computing power of a single robot in the tasks of image processing, target recognition and tracking is often insufficient. A common approach to solve this problem is to use cloud services to allow the robot to obtain additional computing resources from a high-performance cloud. The cloud robot is the combination of cloud computing and robotics, and just like other network terminals, the robot does not need to store all data information and has no super-strong computing power, only needs to put forward a demand to the cloud end, and the cloud end carries out corresponding response and meets the requirements of the robot. The method is characterized in that the robot system is greatly improved in performance by using the powerful data fusion and processing capacity of the cloud service platform. However, in the swarm robot system, such a centralized cloud service method is highly delayed and even unavailable due to factors such as communication link congestion and remote distance. Therefore, a new cloud service mode is needed to realize real-time cloud service support for all robots in the whole group of robot system.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an onboard dynamic cloud system and a real-time response resource allocation method thereof.
The purpose of the invention can be realized by the following technical scheme:
an airborne dynamic cloud system integrates airborne idle computing resources of a plurality of robots to form a dynamic and real-time cloud system, and provides real-time cloud service support for robots in a group of robot systems when executing compute-intensive tasks.
The robots in the system include resource demanding robots and resource providing robots.
And each robot switches between the resource demand robot and the resource supply robot according to the real-time task demand and the idle computing resource condition of the robot.
A real-time response resource allocation method of the onboard dynamic cloud system comprises the following steps:
step S1: a resource demand robot initiates a computation task offloading request;
step S2: according to the maximum principle of available computing resources under the real-time requirement, allocating communication network bandwidth resources according to airborne available computing resources of each robot to obtain an effective bandwidth allocation result;
step S3: and based on the obtained utility bandwidth allocation result and the calculation task, performing task allocation by combining the available calculation resources of each robot.
The step S2 specifically includes:
step S21: determining the signal-to-noise ratio from the resource demand robot to each resource supply robot;
step S22: determining a first intermediate variable based on the signal-to-noise ratio of the resource demand robot to each resource providing robot:
wherein: b is
k(i) Providing a first intermediate variable, k, of the robot for the corresponding ith resource
iProviding the signal-to-noise ratio of the robots from the resource demand robot to the ith resource, wherein n is the number of the robots provided by the resource, and B is the total communication network bandwidth resource;
step S23: determining a transmission rate corresponding to the first intermediate variable;
step S24: judging whether the transmission rate corresponding to the first intermediate variable is smaller than the minimum transmission rate required by the current calculation requirement, if so, taking the first intermediate variable as a second intermediate variable, otherwise, determining the second intermediate variable:
wherein: bv
iProviding a second intermediate variable, Tr, of the robot corresponding to the ith resource
required(i) A minimum transmission rate required for the current computational requirements;
step S24: determining a utility bandwidth allocation result according to the second intermediate variable:
wherein: b is
iAnd the final utility bandwidth allocation result.
The step S3 specifically includes:
step S31: determining an upper time limit meeting the real-time requirement according to the calculation task;
step S32: determining a data volume processable model of the resource demand robot and each resource providing robot;
step S33: and performing task allocation according to the established resource demand robot and the processable data volume model of each resource providing robot.
The processable data volume model of the resource demand robot is as follows:
f
0(S
Data)+t
Map+t
Reduce≤t
Deadline
wherein: f. of
0(S
Data_0) Data processing time for resource demanding robots, S
Data_0Processable data volume for a resource demanding robot, t
MapTime required for data segmentation, t
ReduceFusion time for subtask processing results, t
DeadlineAn upper time limit to meet real-time requirements;
the processable data volume model of the resource providing robot is as follows:
f
i(S
Data_i)+S
Data_i/Tr
i+t
Map+t
Reduce≤t
Deadline
wherein: f. of
i(S
Data_i) Providing the data processing time of the robot for the ith resource, S
Data_iThe ith resource provides a processable data volume, Tr, of the robot
iThe transmission rate between the robot and the resource demanding robot is provided for the ith resource.
The transmission rate between the ith resource providing robot and the resource requiring robot is specifically as follows:
wherein: b is the total bandwidth resource, Tr
required(i) Is the communication transmission rate, Tr, required by the current task
max(i) Is the transmission rate required by the current available computing resources of the robot, k is a system coefficient related to the system state of wavelength, transmitter, receiving antenna, etc., P
iIs the transmission power, d
iIs the distance between the resource providing robot and the resource using robot, and l is the attenuation coefficient of the wireless channel
Compared with the prior art, the invention has the following beneficial effects:
1) the swarm robots are enabled to share computing resources as needed, and a high-performance onboard dynamic cloud is provided for compute-intensive tasks without relying on cloud servers and communication infrastructure.
2) The utility bandwidth and the computing task are combined, so that the maximum effect of available computing resources under the real-time requirement is achieved.
Drawings
FIG. 1 is a schematic diagram of a typical onboard dynamic cloud;
FIG. 2 is a schematic diagram of onboard dynamic cloud computing utility bandwidth allocation;
fig. 3 is a schematic diagram of on-board dynamic cloud task allocation.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
The purpose of taking reports by the equipment is to provide an airborne dynamic cloud and a real-time response resource management method thereof, and the method can integrate airborne computing resources of a plurality of robots within a certain range to form a dynamic and real-time cloud system called as the airborne dynamic cloud for some required robots. Specifically, in the group robot system, if a robot has a complex computing task and airborne idle computing resources cannot meet the computing speed requirement, the robot unloads the computing task to one or more most suitable robots according to the current communication conditions, the idle airborne computing resources of nearby group robots, and other conditions. The robot provides real-time cloud service for the robot through a mechanical dynamic cloud composed of the robots providing computing resources.
However, the hardware configuration of each robot in the group robot system is different, and in addition, tasks with different calculation amounts which are currently executed are added, so that the available calculation resources of each robot are different; at the same time, dynamically changing, limited network resources can result in varying transfer rates being achieved by the robots. These two points present significant difficulties in integrating the available resources in an onboard dynamic cloud. For example, if a device with more idle computing resources allocates less communication resources, it may cause a waste of computing resources; conversely, if a device with less free computing resources allocates more communication resources, it may cause a waste of communication resources. In order to effectively utilize the computing resources in the airborne dynamic cloud, communication and computing resource allocation needs to be reasonably performed according to the current communication network conditions and idle airborne computing resources so as to meet the real-time requirement of tasks.
In order to achieve the above object, an onboard dynamic cloud system is provided, as shown in fig. 1, onboard idle computing resources of a plurality of robots are integrated to form a dynamic and real-time cloud system, so as to provide real-time cloud service support for the robots in a group of robot systems when performing computation-intensive tasks.
The robots in the system include resource demanding robots and resource providing robots.
And each robot switches between the resource demand robot and the resource supply robot according to the real-time task demand and the idle computing resource condition of the robot.
A real-time response resource allocation method of the onboard dynamic cloud system comprises the following steps:
step S1: a resource demand robot initiates a computation task offloading request;
step S2: according to the maximum principle of available computing resources under the real-time requirement, allocating communication network bandwidth resources according to airborne available computing resources of each robot to obtain a utility bandwidth allocation result, wherein the step S2 specifically includes:
step S21: determining the signal-to-noise ratio from the resource demand robot to each resource supply robot;
step S22: determining a first intermediate variable based on the signal-to-noise ratio of the resource demand robot to each resource providing robot:
wherein: b is
k(i) Providing a first intermediate variable, k, of the robot for the corresponding ith resource
iProviding the signal-to-noise ratio of the robots from the resource demand robot to the ith resource, wherein n is the number of the robots provided by the resource, and B is the total communication network bandwidth resource;
step S23: determining a transmission rate corresponding to the first intermediate variable;
step S24: judging whether the transmission rate corresponding to the first intermediate variable is smaller than the minimum transmission rate required by the current calculation requirement, if so, taking the first intermediate variable as a second intermediate variable, otherwise, determining the second intermediate variable:
wherein: bv
iProviding a second intermediate variable, Tr, of the robot corresponding to the ith resource
required(i) A minimum transmission rate required for the current computational requirements;
step S24: determining a utility bandwidth allocation result according to the second intermediate variable:
wherein: b is
iAnd the final utility bandwidth allocation result.
Step S3: and based on the obtained utility bandwidth allocation result and the calculation task, performing task allocation by combining the available calculation resources of each robot.
In response to task allocation in real time, first, a relationship between idle computing resources and a required data transfer rate is identified, for example, the remaining computing capacity of a certain robot can complete processing of 10Kb data within 5ms, and to ensure timeliness, the whole processing process needs to be completed within 10ms, that is, when only data of less than 5ms is transferred, the required data transfer rate needs to be greater than a transmission rate of 2Mb/s which is 10Kb/5 ms; secondly, the size of data which can be processed by the robot in the computer-mounted dynamic cloud on the premise of meeting the real-time requirement is required; and finally, distributing in the most appropriate airborne dynamic cloud according to the data size. In general, i.e.
t
Processing＝f
i(S
Data) (formula 1)
t
Transmision＝S
Data/Tr (formula 2)
In the formula (f)
iRepresents the ith robot, S
DataIs the data size of the required processing, Tr is the data transfer rate, t
ProcessinqIs the data processing time, t
TransmisionIs the data transmission time. Wherein, the size and processing time of the processing data in the formula 1 are related to the specific task type, and the invention is not limited specifically; transfer rate Tr in equation 2 is allocated by the computational utility bandwidthThe method is used for preparing the compound. Definition of t
DeadlineIs the upper time limit, t, that meets the real-time requirement
MapIs the time required for data segmentation, t
ReduceIs the fusion time of the returned subtask processing results. Thus, the amount of data that can be handled by robot i in the onboard dynamic cloud is
f
i(S
Data)+S
Data/Tr+t
Map+t
Reduce≤t
Deadline(formula 3)
It should be noted that the resource demand robot itself also has a part of idle computing resources, and this part of computing resources do not need to perform data transmission, so that the processable data volume of the resource demand robot satisfying the real-time requirement is f
i(S
Data)+t
Map+t
Reduce≤t
Deadline(formula 4)
Therefore, step S3 specifically includes:
step S31: determining an upper time limit meeting the real-time requirement according to the calculation task;
step S32: determining a data volume processable model of the resource demand robot and each resource providing robot;
the processable data volume model of the resource demand robot is as follows:
f
0(S
Data)+t
Map+t
Reduce≤t
Deadline
wherein: f. of
0(S
Data_0) Data processing time for resource demanding robots, S
Data_0 is the processable data volume of the resource demand robot, t
MapTime required for data segmentation, t
ReduceFusion time for subtask processing results, t
DeadlineAn upper time limit to meet real-time requirements;
the processable data volume model of the resource providing robot is as follows:
f
i(S
Data_i)+S
Data_i/Tr
i+t
Map+t
Reduce≤t
Deadline
wherein: f. of
i(S
Data_i) Providing the data processing time of the robot for the ith resource, S
Data_iThe ith resource provides a processable data volume, Tr, of the robot
iThe transmission rate between the robot and the resource demanding robot is provided for the ith resource.
The transmission rate between the ith resource providing robot and the resource demand robot is specifically as follows:
wherein: b is the total bandwidth resource, Tr
required(i) Is the communication transmission rate, Tr, required by the current task
max(i) Is the transmission rate required by the current available computing resources of the robot, k is a system coefficient related to the system state of wavelength, transmitter, receiving antenna, etc., P
iIs the transmission power, d
iIs the distance between the resource providing robot and the resource using robot, and l is the attenuation coefficient of the wireless channel
Step S33: and performing task allocation according to the established resource demand robot and the processable data volume model of each resource providing robot.
Fig. 1 is a schematic diagram of a typical onboard dynamic cloud in an embodiment of the invention. Referring to fig. 1, the onboard dynamic cloud and the real-time response resource management method thereof of the present invention include two types of robots: the robot using the onboard dynamic cloud is the robot using the resources outside the onboard dynamic cloud. Referring to fig. 1, in a group robot system composed of eight robots, two robots have computation requirements for computation-intensive tasks, and six robots have no computation requirements for computation tasks, so that the two resource-using robots provide computation resources for the robots using six resources according to the invention, and the real-time requirements for the computation-intensive tasks are met.
Fig. 2 is a schematic diagram of utility bandwidth allocation of on-board dynamic cloud computing in the embodiment of the present invention. Referring to fig. 2, the onboard dynamic cloud and the real-time response resource management method thereof of the present invention mainly include identifying idle computing resources and allocating limited network resources. The broadcast message of the robot comprises idle computing resource information and current communication resource information of nearby robots. Then, according to the computational utility bandwidth allocation method and the real-time response task allocation method in the onboard dynamic cloud and the real-time response resource management method thereof, real-time task allocation of the onboard dynamic cloud is realized.
Fig. 3 is a schematic diagram of on-board dynamic cloud resource management and task allocation in an embodiment of the present invention. Referring to fig. 3, airborne computing resources of heterogeneous group robots are used to form an airborne dynamic cloud, so as to provide real-time cloud service support for the resource-demanding robots. In the system, if the robot has a calculation requirement, the calculation task is unloaded to the most appropriate robot according to the current communication condition and the available calculation resources of the nearby robot, so that the real-time requirement is met. To achieve this, for a robot with computing requirements, it is first necessary to identify the states of communication resources and idle computing resources in the current system, and then select the most suitable robot among the available resources satisfying the real-time response for the offloading of computing tasks.
Claims (8)
1. An airborne dynamic cloud system is characterized in that airborne idle computing resources of a plurality of robots are integrated to form a dynamic and real-time cloud system, and real-time cloud service support is provided for robots in a group of robot systems when executing computing-intensive tasks.
2. The system of claim 1, wherein the robots in the system comprise resource demanding robots and resource providing robots.
3. The system of claim 2, wherein each robot switches between resource demanding robots and resource providing robots based on its real-time task demands and idle computing resource conditions.
4. A real-time response resource allocation method for an onboard dynamic cloud system according to any one of claims 1 to 3, comprising:
step S1: a resource demand robot initiates a computation task offloading request;
step S2: according to the maximum principle of available computing resources under the real-time requirement, allocating communication network bandwidth resources according to airborne available computing resources of each robot to obtain an effective bandwidth allocation result;
step S3: and based on the obtained utility bandwidth allocation result and the calculation task, performing task allocation by combining the available calculation resources of each robot.
5. The method according to claim 4, wherein the step S2 specifically includes:
step S21: determining the signal-to-noise ratio from the resource demand robot to each resource supply robot;
step S22: determining a first intermediate variable based on the signal-to-noise ratio of the resource demand robot to each resource providing robot:
wherein: b is
k(i) Providing a first intermediate variable, k, of the robot for the corresponding ith resource
iProviding the signal-to-noise ratio of the robots from the resource demand robot to the ith resource, wherein n is the number of the robots provided by the resource, and B is the total communication network bandwidth resource;
step S23: determining a transmission rate corresponding to the first intermediate variable;
step S24: judging whether the transmission rate corresponding to the first intermediate variable is smaller than the minimum transmission rate required by the current calculation requirement, if so, taking the first intermediate variable as a second intermediate variable, otherwise, determining the second intermediate variable:
wherein: bv
iProviding a second intermediate variable, Tr, of the robot corresponding to the ith resource
required(i) A minimum transmission rate required for the current computational requirements;
step S24: determining a utility bandwidth allocation result according to the second intermediate variable:
wherein: b is
iAnd the final utility bandwidth allocation result.
6. The method according to claim 4, wherein the step S3 specifically includes:
step S31: determining an upper time limit meeting the real-time requirement according to the calculation task;
step S32: determining a data volume processable model of the resource demand robot and each resource providing robot;
step S33: and performing task allocation according to the established resource demand robot and the processable data volume model of each resource providing robot.
7. The method of claim 6, wherein the processable data volume model of the resource demanding robot is:
f
0(S
Data)+t
Map+t
Reduce≤t
Deadline
wherein: f. of
0(S
Data_0) Data processing time for resource demanding robots, S
Data_0Processable data volume for a resource demanding robot, t
MapTime required for data segmentation, t
ReduceFusion time for subtask processing results, t
DeadlineAn upper time limit to meet real-time requirements;
the processable data volume model of the resource providing robot is as follows:
f
i(S
Data_i)+S
Data_i/Tr
i+t
Map+t
Reduce≤t
Deadline
wherein: f. of
i(S
Data_i) Providing the data processing time of the robot for the ith resource, S
Data_iThe ith resource provides a processable data volume, Tr, of the robot
iThe transmission rate between the robot and the resource demanding robot is provided for the ith resource.
8. The method according to claim 7, wherein the transmission rate between the ith resource providing robot and the resource demanding robot is specifically:
wherein: b is the total bandwidth resource, Tr
required(i) Is the communication transmission rate, Tr, required by the current task
max(i) Is the transmission rate required by the current available computing resources of the robot, k is a system coefficient related to the system state of wavelength, transmitter, receiving antenna, etc., P
iIs the transmission power, d
iIs the distance between the resource providing robot and the resource using robot, and l is the attenuation coefficient of the wireless channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910878010.8A CN110784512A (en) | 2019-09-17 | 2019-09-17 | Airborne dynamic cloud system and real-time response resource allocation method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910878010.8A CN110784512A (en) | 2019-09-17 | 2019-09-17 | Airborne dynamic cloud system and real-time response resource allocation method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110784512A true CN110784512A (en) | 2020-02-11 |
Family
ID=69383558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910878010.8A Pending CN110784512A (en) | 2019-09-17 | 2019-09-17 | Airborne dynamic cloud system and real-time response resource allocation method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110784512A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113823011A (en) * | 2021-09-03 | 2021-12-21 | 深圳云天励飞技术股份有限公司 | Calculation force distribution method of patrol robot and related equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105630601A (en) * | 2014-11-03 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Resource allocation method and system based on real-time computing |
| US20180048584A1 (en) * | 2016-08-15 | 2018-02-15 | At&T Intellectual Property I, L.P. | Dynamic provisioning of storage in the cloud |
| CN109245933A (en) * | 2018-09-21 | 2019-01-18 | 西北大学 | One kind calculating environmental resource towards cloud combination of edge and finds method |
-
2019
- 2019-09-17 CN CN201910878010.8A patent/CN110784512A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105630601A (en) * | 2014-11-03 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Resource allocation method and system based on real-time computing |
| US20180048584A1 (en) * | 2016-08-15 | 2018-02-15 | At&T Intellectual Property I, L.P. | Dynamic provisioning of storage in the cloud |
| CN109245933A (en) * | 2018-09-21 | 2019-01-18 | 西北大学 | One kind calculating environmental resource towards cloud combination of edge and finds method |
Non-Patent Citations (1)
| Title |
|---|
| QIRONG TANG等: "A Resource Management Algorithm for Real-time Response of Mobile Ad hoc Cloud in Swarm Robotic System", 《PROCEEDINGS OF THE 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113823011A (en) * | 2021-09-03 | 2021-12-21 | 深圳云天励飞技术股份有限公司 | Calculation force distribution method of patrol robot and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109996039B (en) | Target tracking method and device based on edge calculation | |
| CN109862592B (en) | Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation | |
| CN110868700B (en) | Cooperative computing unloading method based on splittable tasks in vehicle-mounted edge computing environment | |
| CN110198339B (en) | QoE (quality of experience) perception-based edge computing task scheduling method | |
| CN108833577B (en) | Data communication optimization system and method based on edge calculation | |
| WO2021023042A1 (en) | Method for searching edge computing server and related device | |
| CN112714164A (en) | Internet of things system and task scheduling method thereof | |
| CN112153697B (en) | CORS (coordinated rotation System) resolving method, broadcasting method and system under multi-base-station high concurrence scene and CORS system | |
| CN113992678A (en) | Calculation migration method for offshore MEC load balancing and resource allocation joint optimization | |
| Yao et al. | Cost-efficient tasks scheduling for smart grid communication network with edge computing system | |
| CN102510403B (en) | Receive and the cluster distributed system and method for real-time analysis for vehicle data | |
| CN109614228B (en) | Comprehensive monitoring front-end system based on dynamic load balancing mode and working method | |
| US9635671B2 (en) | Scheduling method | |
| CN105282045B (en) | A kind of distributed computing and storage method based on consistency hash algorithm | |
| CN110784512A (en) | Airborne dynamic cloud system and real-time response resource allocation method thereof | |
| US20230376355A1 (en) | Methods, Terminals and Network Devices for Computing Task Allocation and Updating | |
| CN114047965A (en) | Computation offloading method, satellite server, and computer-readable storage medium | |
| CN113656150A (en) | Deep learning computing power virtualization system | |
| CN107273082A (en) | Display methods, device, terminal and the storage medium of image | |
| CN112187842B (en) | Local area network data processing system and local area network data processing method | |
| CN111866159A (en) | Method, system, device and storage medium for calling artificial intelligence service | |
| CN113115072A (en) | Video target detection tracking scheduling method and system based on end cloud cooperation | |
| CN107357659B (en) | Grouping method and query method for query of global load balance in Storm continuous range | |
| CN114884945B (en) | Data transmission method, cloud server, device, system and storage medium | |
| CN115981872B (en) | Method and device for calling algorithm resources, electronic equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200211 |