CN114363984B - Cloud edge collaborative optical carrier network spectrum resource allocation method and system - Google Patents
Cloud edge collaborative optical carrier network spectrum resource allocation method and system Download PDFInfo
- Publication number
- CN114363984B CN114363984B CN202111541328.0A CN202111541328A CN114363984B CN 114363984 B CN114363984 B CN 114363984B CN 202111541328 A CN202111541328 A CN 202111541328A CN 114363984 B CN114363984 B CN 114363984B
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- computing
- user request
- edge
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q11/00—Selecting arrangements for multiplex systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
-
- 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
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a method and a system for distributing spectrum resources of a cloud-edge collaborative optical carrier network, which comprise the following steps: s1: reading a cloud edge cooperative optical carrier network topological structure and parameters; s2: generating a user request, and selecting a processing node in the topological structure according to the computing resources required by the user request and the maximum time delay limit; s3: calculating total energy consumption of each path from the user request to the processing node; s4: and selecting a path with the minimum total energy consumption and the idle spectrum slots meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation. The invention effectively balances the relation between resources and service energy consumption, and reduces network energy consumption to the maximum extent under the condition of reasonably distributing resources.
Description
Technical Field
The invention relates to the technical field of optical communication, in particular to a method and a system for distributing spectrum resources of a cloud edge collaborative optical carrier network.
Background
In recent years, with the rapid development of the Internet of Things (IoT) and the diversification of application scenarios of mobile networks, the demand of users for network computing resources has increased dramatically. The computing power of resource-constrained mobile terminal devices has not been able to meet the rapidly increasing demands of mobile users in terms of data processing. While new mobile devices are becoming more powerful at Central Processing Units (CPUs), mobile applications, such as interactive gaming, virtual reality, and natural language Processing, often require intensive computing and mobile devices may not be able to handle such voluminous applications in a short amount of time. Mobile cloud computing has the potential to address the above challenges as a new architecture and technology moves. The mobile cloud computing can provide a large number of mobile application program access functions, and the equipment can transmit computing tasks to the remote cloud server for execution through computing resource unloading, so that the problem of large computing resource demand can be effectively solved. However, the transmission of the computing task to the cloud server causes unacceptable delay and additional transmission energy consumption.
In order to solve the problems of unacceptable high time delay, high energy consumption and the like generated in the unloading process of mobile cloud computing, a large number of novel network architectures and technologies emerge in succession. As early as 2014, the European Telecommunications Standards Institute (ETSI) first proposed the concept of mobile edge computing MEC: a general-purpose server is deployed near the edge of a mobile network, such as a base station, a wireless access site, etc., to provide an IT service environment and cloud computing capabilities for the wireless access network near the user. The MEC is considered as a complementary form of cloud computing, the defects of high delay and high energy consumption caused by insufficient computing resources of a user, limited battery power and transmission of a large amount of data to the cloud computing are overcome, and the technology for providing the mobile user with nearby service delivery and data computing capability becomes one of 5G key technologies. By deploying cloud computing and cloud storage to the edge of the network, a telecommunication service environment with high performance, low time delay and high bandwidth is provided, distribution and downloading of various contents, services and applications in the network are accelerated, and consumers can enjoy higher-quality network experience.
However, compared with cloud computing, the computing resources of the edge server are relatively scarce, and all the sudden task requests cannot be met. The existing research mostly focuses on dispersing the resources of the original cloud data center to the vicinity of the mobile terminal device, and migrating the computing task of the mobile device to the MEC platform to finish the problem of insufficient computing capability of the mobile terminal device. However, the impact of mobile edge computing and cloud data center collaborative computing, and different offloading strategies on system energy consumption and computing resources has not been paid sufficient attention for a long time. On the one hand, the cloud computing center has rich computing resources, but has long transmission time and high energy consumption. On the other hand, edge computing resources are relatively scarce, so that it cannot respond rapidly to the ever-increasing computing demand, and therefore, under a high-load environment, the computing energy consumption of the edge server may exceed that of cloud computing. In order to realize green computing of the computing system, cooperation between the edge server and the cloud server is very important for reducing system energy consumption. In the 5G network cloud edge collaborative computing task scheduling process, in order to guarantee user service quality, not only the computing processing time delay of a mobile edge and a cloud data center but also the transmission time delay from an edge node to the cloud data center need to be considered, and the network resource utilization rate is guaranteed to be improved under the maximum time delay constraint condition of a service request, and the system energy consumption is reduced.
Disclosure of Invention
The invention aims to provide a method and a system for distributing spectrum resources of a cloud-edge collaborative optical carrier network, which can effectively balance the relation between resources and service energy consumption and reduce the network energy consumption to the maximum extent under the condition of reasonably distributing the resources.
In order to solve the technical problem, the invention provides a method for allocating spectrum resources of a cloud-edge collaborative optical carrier network, which comprises the following steps:
s1: reading a cloud edge cooperative optical carrier network topological structure and parameters;
the cloud edge collaborative optical carrier network topology structure comprises network nodes and user requests, the network nodes comprise a base station, an exchanger, an edge computing server and a cloud data center server, and parameters comprise computing resources of each network node and maximum time delay limit met; the user request comprises a frequency spectrum slot, computing resources, data size and maximum time delay limit required by the user request;
s2: generating a user request, and selecting a processing node in the topological structure according to the computing resources required by the user request and the maximum time delay limit;
s3: calculating total energy consumption of each path from a user request to a processing node, specifically comprising:
a. when the computing resources of the edge computing server in the local area meet the computing resources required by the user request, the user request is sent to the local edge computing server for processing, and the total energy consumption comprises sending energy consumption and computing energy consumption of the edge server in the local area;
b. if the computing resources in the local edge computing region are insufficient, judging whether the computing servers in the edge computing region of the adjacent region have enough computing resources to process the user request;
when the computing resources of the edge computing servers of the adjacent areas meet the computing resources required by the user request and the processing time delay of the edge computing servers of the adjacent areas meets the maximum time delay limit required by the user request, the user request is transferred to the edge computing servers of other areas with the minimum processing energy consumption on the same switch in the local area, and the total energy consumption comprises the sending energy consumption, the node energy consumption of the path network nodes and the computing energy consumption of the adjacent edge servers;
c. if the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources or service quality required by the user request, judging whether the cloud data center server has enough computing resources and whether the maximum time delay requirement of the user request is met;
when the cloud data center server has enough computing resources and meets the time delay requirement, a user request is sent to the cloud data center server through the switch for data processing, and at the moment, the total energy consumption comprises sending energy consumption, node energy consumption of the path network nodes and computing energy consumption of the cloud data center server;
s4: selecting a path with the minimum total energy consumption and idle spectrum gaps meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation;
calculating working paths from a user request to a processing node by adopting a K shortest path algorithm, calculating K candidate paths as routing selection, defining the priority of the candidate paths according to the energy consumption and the idle spectrum slot generated by each path, selecting the working paths according to the priority, and defining a formula according to the priority:
wherein S is k And E k Respectively representing the free spectrum slots and the resulting energy consumption, S, of the k-th working path max And S min Representing the maximum and minimum of the free spectrum slots of the K working paths, E max And E min The maximum value and the minimum value of the energy consumption of the K working paths are represented, alpha and beta are two adjustable factors, and alpha + beta =1 is satisfied;
the user requests to calculate the total energy consumption according to the selected processing node, wherein the total energy consumption comprises SEE (energy consumption of sending), NOE (energy consumption of node) and CAE (energy consumption of computing edge server) e And computing energy consumption CAE of cloud data center server c Namely:
wherein p is u Indicating the transmission power of the user equipment, d u Indicating the size of the task, p j Indicating port power, v, of switch j j Representing the forwarding rate of the switch, n representing the number of nodes traversed by the path, delta e Representing the real-time processing capabilities of the edge compute server,
represents the energy consumption, delta, generated by the edge computing server processing the data unit time c Representing the assignment of cloud data center computing servers to each taskProcessing capacity,. L u Representing the number of computing resources requested by the user,
the energy consumption generated by the cloud data center computing server in unit time for processing data is represented;
user request u uplink transmission rate to base station b over radio subchannel:
wherein W represents the channel bandwidth, p u Which represents the transmit power of the user equipment,
indicating the channel gain, σ, of the radio sub-channel used to transmit the user request u to the base station b 2 Representing the noise power.
As a further improvement of the invention, when a path with high priority is blocked on a certain link, paths with lower priority are selected in turn for spectrum resource allocation until the resource allocation is successful or all paths are blocked.
As a further improvement of the invention, the method also comprises the following steps: after the spectrum resources are successfully distributed, real-time updating of computing resources is carried out on processing nodes for processing user requests;
after the user request is transmitted successfully, releasing resources of the frequency spectrum resources occupied by the working path, and simultaneously releasing computing resources of a processing node processing the user request;
and clearing the information of the working path established by the user request.
A cloud edge collaborative optical carrier network spectrum resource allocation system comprises:
the cloud edge collaborative optical carrier network comprises a topological structure and a data processing unit, wherein the topological structure is used for providing computing resources; the topological structure of the cloud edge collaborative optical carrier network comprises network nodes and user requests, wherein the network nodes comprise a base station, an exchanger, an edge computing server and a cloud data center server, and parameters comprise computing resources of each network node and maximum time delay limit met; the user request comprises a frequency spectrum slot, computing resources, data size and maximum time delay limit required by the user request; the user request module is used for generating a user request and selecting a processing node in the topological structure according to the computing resources and the maximum time delay limit required by the user request;
the energy consumption calculation module is used for calculating total energy consumption of each path from the user request to the processing node, and specifically comprises:
a. when the computing resources of the edge computing server in the local area meet the computing resources required by the user request, the user request is sent to the local edge computing server for processing, and the total energy consumption comprises sending energy consumption and computing energy consumption of the edge server in the local area;
b. if the computing resources in the local edge computing region are insufficient, judging whether the edge computing servers in the adjacent region have enough computing resources to process the user request;
when the computing resources of the edge computing servers of the adjacent areas meet the computing resources required by the user request and the processing time delay of the edge computing servers of the adjacent areas meets the maximum time delay limit required by the user request, the user request is transferred to the edge computing servers of other areas with the minimum processing energy consumption on the same switch in the local area, and the total energy consumption comprises the sending energy consumption, the node energy consumption of the path network nodes and the computing energy consumption of the adjacent edge servers;
c. if the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources or service quality required by the user request, judging whether the cloud data center server has enough computing resources and whether the maximum time delay requirement of the user request is met;
when the cloud data center server has enough computing resources and meets the time delay requirement, a user request is sent to the cloud data center server through the switch for data processing, and at the moment, the total energy consumption comprises sending energy consumption, node energy consumption of the path network nodes and computing energy consumption of the cloud data center server;
the path selection and resource allocation module is used for selecting a path with the minimum total energy consumption and idle spectrum gaps meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation;
calculating working paths from a user request to a processing node by adopting a K shortest path algorithm, calculating K candidate paths as routing selection, defining the priority of the candidate paths according to the energy consumption and the idle spectrum slot generated by each path, selecting the working paths according to the priority, and defining a formula according to the priority:
wherein S is k And E k Respectively representing the free spectrum slots and the resulting energy consumption, S, of the k-th working path max And S min Represents the maximum and minimum of the free spectral slots of the K working paths, E max And E min The maximum value and the minimum value of the energy consumption of the K working paths are represented, alpha and beta are two adjustable factors, and alpha + beta =1 is satisfied;
the user requests to calculate the total energy consumption according to the selected processing node, wherein the total energy consumption comprises SEE (energy consumption of sending), NOE (energy consumption of node) and CAE (energy consumption of computing edge server) e And computing energy consumption CAE of cloud data center server c Namely:
wherein p is u Indicating the transmission power of the user equipment, d u Indicating the size of the task, p j Indicating port power, v, of switch j j Indicating the forwarding rate of the switch, n indicating the number of nodes traversed by the path, delta e Representing the real-time processing capabilities of the edge compute server,
represents the energy consumption, delta, generated by the edge computing server processing the data unit time c Representing the processing capacity, l, of the cloud data center computing server to assign to each task u Representing the number of computing resources requested by the user,
the energy consumption generated by the cloud data center computing server in unit time for processing data is represented;
user request u uplink transmission rate to base station b over radio subchannel:
wherein W represents the channel bandwidth, p u Which represents the transmission power of the user equipment,
indicating the channel gain, σ, of the radio sub-channel used to transmit the user request u to the base station b 2 Representing the noise power.
As a further improvement of the present invention, the present invention further includes a network state monitoring module, which is used for monitoring and deciding the cloud-edge collaborative optical carrier network, generating a user request, processing node selection, working path selection, spectrum resource allocation, computing resource update, resource release and energy consumption computation.
The invention has the beneficial effects that: the invention mainly aims at the problem of how to balance computing resources, spectrum resources and service energy consumption, and provides a method and a system for optimizing energy consumption of an optical carrier wireless network of a cloud-edge cooperative network; for each user request, selecting a server for processing the service according to the priority level of the task and the strategy deployed in the service area by the cloud data center and the mobile edge computing; calculating a working path from the service to the edge calculation server by adopting K shortest path algorithms; calculating the priority of a path, selecting a path with high priority to reduce the network blocking rate, after a working path is successfully selected, adopting a first-hit frequency spectrum allocation algorithm to allocate frequency spectrum resources to the path, wherein two constraint conditions of frequency spectrum consistency and frequency spectrum continuity are required to be met simultaneously, and then updating network calculation resources and frequency spectrum resource states in real time; after each user request is successfully established, calculating emission energy consumption according to user sending power, service data size and wireless sub-channel transmission rate, calculating path node energy consumption according to load conditions of the number of nodes passing through a path, the service data size, the node power and the forwarding rate, and calculating the calculated energy consumption of the cloud-edge collaborative optical carrier wireless network according to the calculation capacity of the server and the energy consumption generated by the server in unit time for processing data; by means of the energy consumption optimization method and system for the cloud-edge collaborative optical carrier wireless network, the relation between network resources and service energy consumption is effectively balanced, and the energy consumption of the cloud-edge collaborative optical carrier wireless network is reduced to the maximum extent under the condition that resources are reasonably distributed.
Drawings
FIG. 1 is a schematic overall flow diagram of the process of the present invention;
FIG. 2 is a schematic flow chart of an embodiment of the method of the present invention;
FIG. 3 is a schematic diagram of the system architecture of the present invention;
fig. 4 is a schematic diagram of a cloud-edge cooperative optical carrier network architecture according to an embodiment of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
Referring to fig. 1, the present invention provides a method for allocating spectrum resources of a cloud-edge collaborative optical carrier network, including the following steps:
s1: reading a cloud edge cooperative optical carrier network topological structure and parameters;
s2: generating a user request, and selecting a processing node in the topological structure according to the computing resources required by the user request and the maximum time delay limit;
s3: calculating total energy consumption of each path from the user request to the processing node;
s4: and selecting a path with the minimum total energy consumption and the idle spectrum slots meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation.
According to the method and the device, the unloading and processing of the servers with lower energy consumption are preferentially selected according to different loads of the servers for performing service processing and different energy consumption generated by the service processing in the cloud edge collaborative network architecture, and the resource occupation state of each server is adapted to minimize the energy consumption generated during the service processing. In the invention, three types of energy consumption are mainly considered, including service sending energy consumption, node energy consumption and calculation energy consumption. The service transmission energy consumption refers to energy consumption generated when a user and an edge node of a local service area transmit a service request domain edge server through wireless transmission for processing, and is related to the transmission power of user equipment, the data size of the service request and the wireless transmission rate; the node energy consumption refers to the energy consumption generated when the service passes through the intermediate node in the path in the transmission process, and is related to the number of the passing nodes, the forwarding power and the forwarding rate; the computing energy consumption is related to the computing resource demand of each user and the real-time computing capacity of the server and the unit computing energy consumption of the server, and the cloud data center server and the MEC server have different unit computing energy consumption.
Specifically, as shown in fig. 2, the method for allocating spectrum resources includes the following steps:
1) In the cloud edge collaborative network, initializing the computing resources of the edge computing server and initializing the cloud edge collaborative optical carrier wireless network. In the cloud edge cooperative optical carrier network G (U, B, J, S), wherein U = {1,2, \8230;, U, … U |U| Denotes a set of user requests, B = {1,2, \ 8230;, B, \8230; B |B| Denotes a set of base stations, J = { b +1, b +2, \8230 }, J, \8230j |J| Denotes a set of switches, S = B = {1,2, \8230 }, S, … S |S| Indicating the positions which can be selected by a group of edge computing servers, and considering a cloud computing data center server;
2) A set of user request sets U is generated, each user request U (f, l, d, T) max ) E.g. U, abbreviated as U, where f, l, d, T max Respectively representing the number of spectrum gaps, the number of computing resources, the data size of a service request and the maximum time delay limit required by a user request;
3) For each user request u (f, l, d, T) max ) Firstly, judging whether an edge computing server of a local area requested by a user has enough computing resources, when the computing resources on the edge computing server in the local area meet the computing resources required by the user request, sending the user request to the local edge computing server for processing, wherein the user request does not pass through other nodes in a network, and the system energy consumption only needs to consider the sending energy consumption and the computing energy consumption of the edge server in the local area;
if the computing resources in the local edge computing region are insufficient, judging whether the edge computing servers in the adjacent region have enough computing resources to process the user request;
if the computing resources of the edge computing servers of the adjacent areas meet the computing resources required by the user request, the time delay and the energy consumption for processing the user request are judged, and for the edge computing servers of the adjacent areas with the processing time delay meeting the maximum time delay limit of the user request, the user request needs to be transferred to the edge computing servers of other areas with the minimum processing energy consumption on the same switch in the local area. At the moment, the system energy consumption comprises user request sending energy consumption, node energy consumption generated when the user requests to access network nodes such as an exchanger and the like, and computing energy consumption of adjacent edge servers;
if the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources or service quality required by the user request, whether the cloud data center server has enough computing resources and whether the maximum delay requirement is met is judged. And if the cloud server has enough computing resources and meets the time delay requirement, the user request is sent to the cloud data center server through the switch for data processing. At the moment, the system energy consumption comprises the sending energy consumption of the user request, the node energy consumption and the computing energy consumption of the cloud server;
4) And calculating a working path from the user request to the edge calculation server by adopting K shortest path algorithms. K candidate paths are calculated by the K shortest path algorithm to be used as route selection, the priority of the candidate paths is defined according to the energy consumption and the idle spectrum slot generated by each path, and the path selection right with higher priority is higher. When the path with high priority is blocked on a certain link, the paths with lower priority are selected in turn for spectrum resource allocation until the resource allocation is successful or all paths are blocked. The priority definition formula is as formula (1):
wherein S is k And E k Respectively representing the free spectrum slots and the resulting energy consumption, S, of the k-th working path max And S min Representing the maximum and minimum of the free spectrum slots of the K working paths, E max And E min The maximum value and the minimum value of the energy consumption of the K working paths are represented, alpha and beta are two adjustable factors, and alpha + beta =1 is satisfied.
5) The specific energy consumption calculation method comprises the following steps:
respectively calculating the sending energy consumption SEE and the node energy consumption NOE of the K working paths according to formulas (2) and (3); for energy consumption calculation of the cloud edge collaborative optical carrier wireless network, processing in different edge calculation servers according to user requests for calculation:
1) If the user requests to process data in the edge computing server, the CAE is calculated according to the formula (4) e ；
2) If the user requests to perform data processing on the cloud data center computing server, the CAE is obtained according to the formula (5) c . According to the principle of minimum energy consumption,selecting a suitable work path to suggest a user request:
wherein p is u Indicating the transmission power of the user equipment, d u Indicating the size of the task, p j Indicating port power, v, of switch j j Representing the forwarding rate of the switch, n representing the number of nodes traversed by the path, delta e Representing the real-time processing capabilities of the edge compute server,
represents the energy consumption of the edge computing server in processing data unit time, delta c Representing the processing capacity, l, of the cloud data center computing server to assign to each task u Representing the number of computing resources requested by the user,
the energy consumption generated by the cloud data center computing server for processing the data unit time is represented.
Further, for each user request u (f, l, d, T) that cannot be computed at the user device itself max ) And sending the information to the base station responsible for the application service in the area. User request u (f, l, d, T) max ) The uplink transmission rate transmitted to the base station b through the radio sub-channel is defined as formula (6) according to the fragrance formula:
wherein W represents the channel bandwidth, p u Which represents the transmit power of the user equipment,
indicating the channel gain, σ, of the radio sub-channel used to transmit the user request u to the base station b 2 Representing the noise power.
6) User request u (f, l, d, T) max ) And after the working path is successfully established, performing spectrum resource allocation on the working path according to the constraint conditions of spectrum consistency and spectrum continuity. Here, a first-hit spectrum allocation algorithm is adopted, a spectrum resource table is generated according to the spectrum resource states of all links on a path for numbering, and an available spectrum gap is searched from the end with a small number. If the available spectrum gap is found, performing spectrum resource allocation and performing spectrum state updating; if not, the spectrum allocation fails and the service is blocked.
Further, the user requests u (f, l, d, T) max ) After a working path is established and spectrum resources are distributed, updating the computing resources of the edge computing server, and recording the number of successfully established user requests; and updating the cloud edge collaborative radio over fiber network state, and repeating the steps 2) -6) for the rest user requests.
As shown in fig. 3, the present invention further provides a system for allocating spectrum resources of a cloud-edge collaborative optical carrier network, including:
the cloud edge collaborative optical carrier network comprises a topological structure and a data processing module, wherein the topological structure is used for providing computing resources;
the user request module is used for generating a user request and selecting a processing node in the topological structure according to the computing resources and the maximum time delay limit required by the user request;
the energy consumption calculation module is used for calculating the total energy consumption of each path from the user request to the processing node;
and the path selection and resource allocation module is used for selecting a path with the minimum total energy consumption and the idle spectrum gaps meeting the constraint conditions of spectrum consistency and continuity as a working path and allocating spectrum resources.
Specifically, in a cloud-edge cooperative network G (U, B, J, S), topology information of a network, a network connection state, the number of user requests, the number of edge computing servers, and the number of base stations and switches are configured; the user request module generates a group of user requests according to the user requests, configures the information such as the number of the user requests, the number of frequency spectrum gaps required by different user requests, computing resources, data size, maximum time delay limit and the like, and selects processing nodes:
firstly, whether an edge computing server of a local area requested by a user has computing resources required by the user request and meets the maximum time delay limit of the edge computing server is judged, and if the computing resources of the local edge computing server are enough and the time delay is small, the user request is directly processed locally. If the local server has insufficient computing resources or the waiting time is too long, considering whether the edge computing servers of other areas outside the local area have the computing resources required by the user request, if the servers of other areas have sufficient computing resources and meet the delay requirement, migrating the user request to other areas through the switch for processing, and selecting the edge computing server of the adjacent area with the minimum processing energy consumption. And if the computing resources of the local area and other areas do not meet the computing resources of the user request, migrating the user request to a cloud data center server through a switch for processing.
The system further comprises: the working path establishing module: according to the user request u (f, l, d, T) max ) The user request and the server for processing the request adopt K shortest path algorithms to calculate K candidate paths from the user request to the server so as to find out the optimal path as a working path.
According to K candidate paths of the working path establishing module, the energy consumption calculating module calculates the user sending energy consumption of each working path, the energy consumption delay of the path nodes and the total energy consumption, and the specific calculating mode is as follows: after each user request successfully establishes the request, recording each user requestAnd solving node information passed by the transmission path, calculating user sending energy consumption and node energy consumption of each user request according to formulas (2) and (3), recording the position of the edge calculation server determined by the user request, and calculating the calculated energy consumption of each user request by using a formula (4) or (5). Then, the idle frequency spectrum slot numbers of K paths are respectively counted, the path selection and resource allocation module selects the working path with the highest priority for service processing transmission and processing according to a path priority definition formula, and the working path with the highest priority is selected according to a user request u (f, l, d, T) max ) Searching bandwidth resources required by meeting the user request in the selected working path according to the required frequency spectrum gap number f, and if dual constraint conditions of frequency spectrum continuity and frequency spectrum consistency are met at the same time, successfully establishing the user request; and if the dual constraint conditions of the spectrum continuity and the spectrum consistency cannot be simultaneously met, the user request establishment fails.
The system further comprises: a computing resource update module: after the spectrum resources are successfully distributed, updating the computing resources of the edge computing server for processing the user request in real time;
a resource release module: after the user request is transmitted successfully, the resource release is carried out on the frequency spectrum resource occupied by the working path, meanwhile, the computing resource of the edge computing server which processes the user request is released, and finally, the information of the working path established by the user request is cleared.
The network state monitoring module is mainly used for completing the state monitoring functions of initialization of the cloud-edge collaborative optical carrier wireless network, user request generation, service priority selection, edge computing server selection, working path establishment, frequency spectrum resource allocation, computing resource updating, resource release and network energy consumption calculation so as to achieve the aim of reducing the system energy consumption as much as possible during the computing of resource allocation; and executing the coordination function among the modules, and judging and early warning whether each module is established successfully, thereby finishing the aim of reducing the energy consumption of the system in the mobile edge calculation.
Examples
As shown in fig. 4, an architecture diagram of a cloud-edge cooperative radio over fiber network includes a cloud computing center, two switches connected to 8 local areas, each area including a base station and a corresponding edge server, and the coverage area of different base stations connected to the same switch is an adjacent area. The computing resources of the edge computing server in the area covered by each base station are assumed to be 20, the computing resources of the cloud data center are 1000, the size of service data is randomly generated within 200-800 Kb, the maximum time delay is limited within 0.5-1 s, and the number of the computing resources, the number of frequency spectrum slots and the number of the frequency spectrum slots are randomly generated within a reasonable range.
Firstly, initializing a cloud-edge cooperative radio over fiber network G (U, B, J, S), including a user request, a base station, a switch, and a position where an edge computing server can select, and initializing computing resources of the edge computing server. U for user request i (f,l,d,T max ) Denotes u i Representing the number of the user request, f representing the number of spectrum slots required for establishing the working path, l representing the computing resource required by the user request, d representing the data size of the user request, and T max Representing the maximum latency requirement requested by the user. Generating 3 user request sets u in node 1 and node 6 base station areas in FIG. 3 1 (3,15,200,0.6)、u 2 (8,10,300,0.8)、 u 3 (5,30,500,1)。
Second, for user requests u generated by node 1 1 (3, 15,200, 0.6), it is first determined whether the user request can be processed locally. Since the number of computing resources requested by the user is 15, and the computing resources of the edge computing server node 2 in the local area are 20, it is assumed that the time delay for the edge computing server in the local area to process the user request is 0.3s at this time. At this time, the user request is sent to the local edge computing server for processing, and the data transmission route is (1).
Third, for user requests u generated by node 1 2 (8, 10,300, 0.8), since the dynamic computing resource of the node 2 at this time is 5, the computing resource required by the user request cannot be met, the computing resource required by the user needs to be migrated to the adjacent edge computing servers of other areas, such as the node 3, the node 4 and the node 5, through the switch node 8, if the computing result is unloaded to the node 3, the node 4 and the node 5, the energy consumption relationship is E 3 >E 5 >E 4 (E 3 、E 4 、E 5 Representing the energy consumption of the nodes 3, 4, 5, respectively), the corresponding time delays are 0.6s,0.9s,0.7s, respectively, at which time the computing resources required by the user are migrated to the node 4 on the adjacent other region edge computing server, the computing resources of the node 4 are updated to 10, and the transmission path is (2).
Fourth, user request u generated for node 6 3 (5, 30,500 and 1), at this time, the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources required by the user request, assuming that the processing delay of the user request unloaded to the cloud data center is calculated to be 0.8s, the maximum delay requirement is met, the user request needs to be transmitted to a cloud data center computing server node 9 connected with the switch through a switch node 7, and the data transmission route is (3).
Fifth, for user request u 1 (3,15,200,0.6)、u 2 (8,10,300,0.8)、u 3 (5, 30,500, 1) K shortest paths are needed to be adopted for calculating K paths from the node 1 to the node 2, from the node 1 to the node 4 and from the node 6 to the node 9 respectively. Respectively calculating the path transmission energy consumption, the node energy consumption and the total calculation energy consumption of the K paths according to a formula (2), a formula (3), a formula (4) and a formula (5), calculating the priorities of the K paths according to a formula (1), and selecting one path with the highest priority from the K paths as a working path.
Sixthly, after the working path is selected, the first hit frequency spectrum allocation algorithm is adopted, and the user request u is requested according to the constraint conditions of frequency spectrum consistency and frequency spectrum continuity 1 (3,15,200,0.6)、u 2 (8,10,300,0.8)、u 3 And (5, 30,500, 1) carrying out spectrum resource allocation on the working path. After the spectrum resources are distributed, a user requests to be successfully established, and at the moment, the states of the computing resources and the spectrum resources are updated in real time and the connection success number is recorded.
Therefore, the invention provides a cloud-edge coordinated optical wireless network energy consumption optimization method and system taking the guarantee of network service quality as a prerequisite and taking the time delay and the resource into consideration comprehensively. According to different loads of the MEC servers, the service requests are unloaded to the server with the minimum energy consumption, so that the energy consumption of the system is reduced to the maximum extent; meanwhile, the spectrum resources in the network also need to be considered, so as to further improve the service quality of the network.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (5)
1. A method for distributing spectrum resources of a cloud edge collaborative optical carrier network is characterized by comprising the following steps: the method comprises the following steps:
s1: reading a cloud edge cooperative optical carrier network topological structure and parameters;
the cloud edge collaborative optical carrier network topology structure comprises network nodes and user requests, the network nodes comprise a base station, an exchanger, an edge computing server and a cloud data center server, and parameters comprise computing resources of each network node and maximum time delay limit met; the user request comprises a frequency spectrum slot, computing resources, data size and maximum time delay limit required by the user request;
s2: generating a user request, and selecting a processing node in a topological structure according to the computing resource required by the user request and the maximum time delay limit; s3: calculating total energy consumption of each path from a user request to a processing node, specifically comprising:
a. when the computing resources of the edge computing server in the local area meet the computing resources required by the user request, the user request is sent to the local edge computing server for processing, and the total energy consumption comprises sending energy consumption and computing energy consumption of the edge server in the local area;
b. if the computing resources in the local edge computing region are insufficient, judging whether the edge computing servers in the adjacent region have enough computing resources to process the user request;
when the computing resources of the edge computing servers of the adjacent areas meet the computing resources required by the user request and the processing time delay of the edge computing servers of the adjacent areas meets the maximum time delay limit required by the user request, the user request is transferred to the edge computing servers of other areas with the minimum processing energy consumption on the same switch in the local area, and the total energy consumption comprises the sending energy consumption, the node energy consumption of the path network nodes and the computing energy consumption of the adjacent edge servers;
c. if the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources or service quality required by the user request, judging whether the cloud data center server has enough computing resources and whether the maximum time delay requirement of the user request is met;
when the cloud data center server has enough computing resources and meets the time delay requirement, a user request is sent to the cloud data center server through the switch for data processing, and at the moment, the total energy consumption comprises sending energy consumption, node energy consumption of the path network nodes and computing energy consumption of the cloud data center server;
s4: selecting a path with the minimum total energy consumption and idle spectrum gaps meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation;
calculating working paths from a user request to a processing node by adopting a K shortest path algorithm, calculating K candidate paths as routing selection, defining the priority of the candidate paths according to the energy consumption and the idle spectrum slot generated by each path, selecting the working paths according to the priority, and defining a formula according to the priority:
wherein S is k And E k Respectively representing the free spectrum slots and the resulting energy consumption, S, of the k-th working path max And S min Indicating that K work paths are emptyMaximum and minimum of idle spectral slots, E max And E min The maximum value and the minimum value of the energy consumption of the K working paths are represented, alpha and beta are two adjustable factors, and alpha + beta =1 is satisfied;
the user requests to calculate the total energy consumption according to the selected processing node, wherein the total energy consumption comprises SEE (energy consumption of sending), NOE (energy consumption of node) and CAE (energy consumption of computing edge server) e And computing energy consumption CAE of cloud data center server c Namely:
wherein p is u Indicating the transmission power of the user equipment, d u Indicating the size of the task, p j Indicating port power, v, of switch j j Indicating the forwarding rate of the switch, n indicating the number of nodes traversed by the path, delta e Representing the real-time processing power of the edge compute server,
represents the energy consumption of the edge computing server in processing data unit time, delta c Representing the processing capacity, l, of the cloud data center computing server to assign to each task u Representing the number of computing resources requested by the user,
the energy consumption generated by the cloud data center computing server in unit time for processing data is represented;
user request u uplink transmission rate to base station b over radio subchannel:
wherein W represents the channel bandwidth, p u Which represents the transmit power of the user equipment,
indicating the channel gain, σ, of the radio sub-channel used to transmit the user request u to the base station b 2 Representing the noise power.
2. The method for allocating spectrum resources in a cloud-edge collaborative optical carrier network according to claim 1, wherein: when the path with high priority is blocked on a certain link, the paths with lower priority are selected in turn for spectrum resource allocation until the resource allocation is successful or all paths are blocked.
3. The method for allocating spectrum resources of a cloud-edge collaborative optical carrier network according to any one of claims 1-2, wherein: further comprising the steps of: after the spectrum resources are successfully distributed, real-time updating of computing resources is carried out on processing nodes which process user requests;
after the user request is transmitted successfully, releasing resources of the frequency spectrum resources occupied by the working path, and simultaneously releasing computing resources of a processing node processing the user request;
and clearing the information of the working path established by the user request.
4. A cloud edge collaborative optical carrier network spectrum resource allocation system is characterized in that: the method comprises the following steps:
the cloud edge collaborative optical carrier network comprises a topological structure and a data processing module, wherein the topological structure is used for providing computing resources; the topological structure of the cloud edge collaborative optical carrier network comprises network nodes and user requests, wherein the network nodes comprise a base station, an exchanger, an edge computing server and a cloud data center server, and parameters comprise computing resources of each network node and maximum time delay limit met; the user request comprises a frequency spectrum slot, computing resources, data size and maximum time delay limit required by the user request;
the user request module is used for generating a user request and selecting a processing node in the topological structure according to the computing resources and the maximum time delay limit required by the user request;
the energy consumption calculation module is used for calculating total energy consumption of each path from the user request to the processing node, and specifically comprises:
a. when the computing resources of the edge computing server in the local area meet the computing resources required by the user request, the user request is sent to the local edge computing server for processing, and the total energy consumption comprises sending energy consumption and computing energy consumption of the edge computing server in the local area;
b. if the computing resources in the local edge computing region are insufficient, judging whether the computing servers in the edge computing region of the adjacent region have enough computing resources to process the user request;
when the computing resources of the edge computing servers of the adjacent areas meet the computing resources required by the user request and the processing time delay of the edge computing servers of the adjacent areas meets the maximum time delay limit required by the user request, the user request is transferred to the edge computing servers of other areas with the minimum processing energy consumption on the same switch in the local area, and the total energy consumption comprises the sending energy consumption, the node energy consumption of the network nodes and the computing energy consumption of the adjacent edge servers;
c. if the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources or service quality required by the user request, judging whether the cloud data center server has enough computing resources and meets the maximum time delay requirement of the user request;
when the cloud data center server has enough computing resources and meets the time delay requirement, the user request is sent to the cloud data center server through the switch for data processing, and at the moment, the total energy consumption comprises sending energy consumption, node energy consumption of the path network nodes and computing energy consumption of the cloud data center server;
the path selection and resource allocation module is used for selecting a path with the minimum total energy consumption and idle spectrum gaps meeting the constraint conditions of spectrum consistency and continuity as a working path and performing spectrum resource allocation;
calculating working paths from a user request to a processing node by adopting a K shortest path algorithm, calculating K candidate paths as routing selection, defining the priority of the candidate paths according to the energy consumption and the idle spectrum slot generated by each path, selecting the working paths according to the priority, and defining a formula according to the priority:
wherein S is k And E k Respectively representing the free spectrum slots and the resulting energy consumption, S, of the k-th working path max And S min Representing the maximum and minimum of the free spectrum slots of the K working paths, E max And E min The maximum value and the minimum value of the energy consumption of the K working paths are represented, alpha and beta are two adjustable factors, and alpha + beta =1 is satisfied;
the user requests to calculate the total energy consumption according to the selected processing node, wherein the total energy consumption comprises SEE (energy consumption of sending), NOE (energy consumption of node) and CAE (energy consumption of computing edge server) e And computing energy consumption CAE of cloud data center server c Namely:
wherein p is u Indicating the transmission power of the user equipment, d u Indicating the size of the task, p j Indicating port power, v, of switch j j Representing the forwarding rate of the switch, n representing the number of nodes traversed by the path, delta e Representing the real-time processing capabilities of the edge compute server,
represents the energy consumption of the edge computing server in processing data unit time, delta c Representing the processing capacity, l, of the cloud data center computing server to assign to each task u Representing the number of computing resources requested by the user,
the energy consumption generated by the cloud data center computing server in unit time for processing data is represented;
uplink transmission rate of user request u to base station b via radio subchannel:
wherein W represents the channel bandwidth, p u Which represents the transmit power of the user equipment,
indicating the channel gain, σ, of the radio sub-channel used to transmit the user request u to the base station b 2 Representing the noise power.
5. The system according to claim 4, wherein the system for allocating spectrum resources in the cloud-edge collaborative optical network over fiber is characterized in that: the system also comprises a network state monitoring module which is used for monitoring and judging the cloud edge collaborative optical carrier network, generating a user request, processing node selection, working path selection, spectrum resource allocation, computing resource updating, resource releasing and energy consumption computing.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111541328.0A CN114363984B (en) | 2021-12-16 | 2021-12-16 | Cloud edge collaborative optical carrier network spectrum resource allocation method and system |
| PCT/CN2021/140060 WO2023108718A1 (en) | 2021-12-16 | 2021-12-21 | Spectrum resource allocation method and system for cloud-edge collaborative optical carrier network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111541328.0A CN114363984B (en) | 2021-12-16 | 2021-12-16 | Cloud edge collaborative optical carrier network spectrum resource allocation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114363984A CN114363984A (en) | 2022-04-15 |
| CN114363984B true CN114363984B (en) | 2022-11-25 |
Family
ID=81098595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111541328.0A Active CN114363984B (en) | 2021-12-16 | 2021-12-16 | Cloud edge collaborative optical carrier network spectrum resource allocation method and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114363984B (en) |
| WO (1) | WO2023108718A1 (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024049334A1 (en) * | 2022-08-30 | 2024-03-07 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and system for energy efficient service placement in an edge cloud |
| CN116680625B (en) * | 2023-08-04 | 2024-01-05 | 山东华科信息技术有限公司 | Cloud edge end cooperation-based distribution network multi-scene matching data processing method and system |
| CN116828226B (en) * | 2023-08-28 | 2023-11-10 | 南京邮电大学 | Cloud edge end collaborative video stream caching system based on block chain |
| CN117255368B (en) * | 2023-11-17 | 2024-02-27 | 广东工业大学 | Edge dynamic integration method for vehicle-mounted edge server and cooperative fixed edge server |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111901424A (en) * | 2020-07-28 | 2020-11-06 | 苏州大学 | Cloud edge cooperative network resource smooth migration and reconstruction method and system |
| CN113364850A (en) * | 2021-06-01 | 2021-09-07 | 苏州路之遥科技股份有限公司 | Software-defined cloud-edge collaborative network energy consumption optimization method and system |
| CN113784373A (en) * | 2021-08-24 | 2021-12-10 | 苏州大学 | Combined optimization method and system for time delay and frequency spectrum occupation in cloud edge cooperative network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9525726B2 (en) * | 2013-06-24 | 2016-12-20 | Nec Corporation | Compute followed by network load balancing procedure for embedding cloud services in software-defined flexible-grid optical transport networks |
| CN110414373B (en) * | 2019-07-08 | 2021-09-24 | 武汉大学 | Deep learning palm vein recognition system and method based on cloud edge-side cooperative computing |
| CN113742046A (en) * | 2021-09-17 | 2021-12-03 | 苏州大学 | Flow grooming cloud-side computing network computing resource balanced scheduling method and system |
-
2021
- 2021-12-16 CN CN202111541328.0A patent/CN114363984B/en active Active
- 2021-12-21 WO PCT/CN2021/140060 patent/WO2023108718A1/en unknown
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111901424A (en) * | 2020-07-28 | 2020-11-06 | 苏州大学 | Cloud edge cooperative network resource smooth migration and reconstruction method and system |
| CN113364850A (en) * | 2021-06-01 | 2021-09-07 | 苏州路之遥科技股份有限公司 | Software-defined cloud-edge collaborative network energy consumption optimization method and system |
| CN113784373A (en) * | 2021-08-24 | 2021-12-10 | 苏州大学 | Combined optimization method and system for time delay and frequency spectrum occupation in cloud edge cooperative network |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023108718A1 (en) | 2023-06-22 |
| CN114363984A (en) | 2022-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114363984B (en) | Cloud edge collaborative optical carrier network spectrum resource allocation method and system | |
| CN111901424B (en) | Cloud edge cooperative network resource smooth migration and reconstruction method and system | |
| CN109413724B (en) | MEC-based task unloading and resource allocation scheme | |
| Nassar et al. | Reinforcement learning for adaptive resource allocation in fog RAN for IoT with heterogeneous latency requirements | |
| CN113364850B (en) | Software-defined cloud-edge collaborative network energy consumption optimization method and system | |
| Xia et al. | Data, user and power allocations for caching in multi-access edge computing | |
| WO2023039965A1 (en) | Cloud-edge computing network computational resource balancing and scheduling method for traffic grooming, and system | |
| WO2023024219A1 (en) | Joint optimization method and system for delay and spectrum occupancy in cloud-edge collaborative network | |
| CN104737609A (en) | System and method for radio access virtualization | |
| CN107911856B (en) | Separation multi-access method based on matching game in ultra-dense heterogeneous network | |
| US20150032495A1 (en) | System and Method for User Controlled Cost Based Network and Path Selection across Multiple Networks | |
| Zhang et al. | DMRA: A decentralized resource allocation scheme for multi-SP mobile edge computing | |
| CN109947574A (en) | A kind of vehicle big data calculating discharging method based on mist network | |
| CN115396514B (en) | Resource allocation method, device and storage medium | |
| Gu et al. | Context-aware task offloading for multi-access edge computing: Matching with externalities | |
| WO2021003422A1 (en) | Network and method for servicing a computation request | |
| CN112492687A (en) | Adaptive resource allocation method and system based on wireless network slice | |
| Huang et al. | Distributed resource allocation for network slicing of bandwidth and computational resource | |
| Gong | Optimal edge server and service placement in mobile edge computing | |
| Shafigh et al. | Dynamic network slicing for flexible radio access in tactile internet | |
| TWI826552B (en) | Spectrum management equipment, electronic equipment, wireless communication method and storage medium | |
| KR102268470B1 (en) | Method and apparatus for configuring virtual network | |
| JP2022503762A (en) | Cellular telecommunications network | |
| CN112867088B (en) | Dynamic adjustment method and system for cloud-edge-end communication architecture | |
| Noghani et al. | A generic framework for task offloading in mmWave MEC backhaul networks |
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 |