CN117459587B - Scheduling method of content distribution network based on edge calculation - Google Patents
Scheduling method of content distribution network based on edge calculation Download PDFInfo
- Publication number
- CN117459587B CN117459587B CN202311778147.9A CN202311778147A CN117459587B CN 117459587 B CN117459587 B CN 117459587B CN 202311778147 A CN202311778147 A CN 202311778147A CN 117459587 B CN117459587 B CN 117459587B
- Authority
- CN
- China
- Prior art keywords
- edge
- content
- node
- user
- nodes
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000004364 calculation method Methods 0.000 title claims abstract description 19
- 238000002955 isolation Methods 0.000 claims abstract description 20
- 230000002159 abnormal effect Effects 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000012544 monitoring process Methods 0.000 claims abstract description 10
- 230000005856 abnormality Effects 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 28
- 238000007667 floating Methods 0.000 claims description 16
- 238000012163 sequencing technique Methods 0.000 claims description 11
- 238000012384 transportation and delivery Methods 0.000 claims description 11
- 239000006185 dispersion Substances 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 3
- 239000002699 waste material Substances 0.000 abstract description 3
- 238000004458 analytical method Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013024 troubleshooting Methods 0.000 description 2
- 238000012550 audit Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011010 flushing procedure Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
-
- 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
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a scheduling method of a content distribution network based on edge calculation, which relates to the technical field of network scheduling, wherein a monitoring end monitors performance data of all edge nodes in real time, judges whether the edge nodes are abnormal after analyzing the performance data of the edge nodes, a processing end marks the edge nodes with the abnormality into an isolation area and stops using, marks all the edge nodes with the distance from a user being smaller than a distance threshold into a to-be-selected area, acquires multiple data of request content, comprehensively analyzes the performance data of the edge nodes, the multiple data of the request content and the user data through edge calculation equipment at the to-be-selected area, sequentially sorts the edge nodes in the to-be-selected area, generates a node sorting table, and caches the request content onto the edge nodes with first sorting in the node sorting table. When the request content cache allocation is carried out, the method evaluates the edge nodes, so that the node resource waste is avoided, and the stability of the request content cache is ensured.
Description
Technical Field
The invention relates to the technical field of network scheduling, in particular to a scheduling method of a content distribution network based on edge calculation.
Background
The content delivery network (Edge Content Delivery Network, edge CDN) is a network architecture that is widely used to improve website performance and content delivery. The method and the system improve the content delivery speed and availability by caching the content on servers in all the world and routing the user request to the nearest server;
the Edge computing based content delivery network (Edge Content Delivery Network, edge CDN) is an emerging network architecture aimed at improving the efficiency and performance of content delivery, which combines the advantages of Edge computing and traditional content delivery networks, pushing content caching and processing towards the network Edge to reduce latency, improve availability and reduce bandwidth consumption, where scheduling methods play a critical role as they determine which content should be cached at the Edge nodes, and how to dynamically route user requests to the Edge nodes.
Existing scheduling methods typically cache content on edge nodes closer to the user, without evaluation processing on the edge nodes, which can lead to the following problems:
caching content on edge nodes closest to the user may lead to a problem of unbalanced load, some edge nodes may take on more request traffic, while other nodes may be relatively idle, which may lead to some nodes being overloaded, while some node resources are wasted;
if content is cached based on user location alone, without regard to the popularity distribution of the content, it may result in some content becoming very popular on some nodes and other content being rarely accessed on other nodes, which may lead to wasted resources and performance issues.
Disclosure of Invention
The invention aims to provide a scheduling method of a content distribution network based on edge calculation, which aims to solve the defects in the background technology.
In order to achieve the above object, the present invention provides the following technical solutions: a method of scheduling a content distribution network based on edge computation, the method comprising the steps of:
s1: the monitoring end monitors the performance data of all the edge nodes in real time, and judges whether the edge nodes are abnormal or not after analyzing the performance data of the edge nodes;
s2: the processing end marks the edge node with abnormality into an isolation area and stops using, and isolation area information is sent to a network administrator;
s3: capturing and analyzing a user request, and scribing all edge nodes with the distance from the user being smaller than a distance threshold value into a candidate area;
s4: acquiring multiple items of data of the request content, comprehensively analyzing the performance data of the edge nodes, the multiple items of data of the request content and the user data through edge computing equipment at the area to be selected, and sequentially sequencing the edge nodes in the area to be selected to generate a node sequencing table;
s5: caching the request content to the first edge node in the node sorting table;
s6: and dynamically adjusting the strategy according to the real-time network condition of the edge node and the user demand.
Preferably, in step S4, a plurality of pieces of data of the requested content are obtained, and performance data of the edge node, a plurality of pieces of data of the requested content, and user data are comprehensively analyzed by the edge computing device at the candidate area;
the performance data comprises node load deviation rate and storage capacity discrete degree at sampling time points, the plurality of data of the requested content comprises content estimated capacity index, and the user data comprises user flow floating coefficient.
Preferably, in step S4, the edge nodes in the to-be-selected area are sequentially ordered, and the generating the node ordering table includes the following steps:
s4.1: comprehensively calculating node load deviation rate, sampling time point storage capacity discrete degree, content estimated capacity index and user flow floating coefficient by using edge computing equipment at a candidate area to obtain a sequencing coefficientThe computational expression is: />Wherein->Estimating a capacity index for content->Floating coefficient for user traffic, +.>For the node load deviation rate, < >>For the degree of dispersion of the storage capacity at the sampling time point +.>、/>、/>、/>The content estimated capacity index, the user flow floating coefficient, the node load deviation rate and the proportional coefficient of the storage capacity discrete degree at the sampling time point are respectively, and +.>、/>、/>、/>Are all greater than 0;
s4.2: obtaining ranking coefficientsAfter the value, all edge nodes in the candidate area are added according to the sorting coefficient>And sorting the values from big to small to generate a node sorting table.
Preferably, the user flow rate floating coefficientThe calculated expression of (2) is:，/>the user traffic real-time variable is requested for the edge node content,requesting a period of user login for edge node content, < >>Requesting a user log-out period for edge node content.
Preferably, the content predictive capacity indexThe calculated expression of (2) is:wherein->Memory occupied by the content itself>Memory occupied for content history average access +.>Is the rated capacity of the edge node.
Preferably, the logic for acquiring the memory occupied by the content history average access is:
acquiring the memory occupied by user request nodes in different time periods when the content is released from the last edge node;
and after acquiring the occupied memory of the user request nodes in all the time periods, calculating an average value to obtain the memory occupied by the content history average access.
Preferably, the calculation expression of the node load deviation rate is:wherein->For the node load deviation rate, < >>For the rated load of the edge node, +.>Is the real-time load of the edge node.
Preferably, the calculation expression of the discrete degree of the storage capacity at the sampling time point is:in the formula->，/>Represents the number of sampling time points, +.>Is a positive integer>Representing the remaining storage capacity of the edge node acquired at the ith sampling time point, +.>Representing the average value of the remaining storage capacity of the edge node.
Preferably, when the node load deviation rate is greater than a deviation threshold value or the storage capacity discrete degree of the sampling time point is greater than a discrete threshold value, the edge node is analyzed to have abnormality.
In the technical scheme, the invention has the technical effects and advantages that:
according to the method, the monitoring end monitors the performance data of all edge nodes in real time, after analyzing the performance data of the edge nodes, judging whether the edge nodes are abnormal, and the processing end marks the abnormal edge nodes into an isolation area and stops using, so that the safety and stability of the request content cache are guaranteed, user requests are captured and analyzed, all the edge nodes with the distance from the user being smaller than a distance threshold value are marked into a to-be-selected area, multiple data of the request content are obtained, after comprehensively analyzing the performance data of the edge nodes, multiple data of the request content and the user data through edge computing equipment at the to-be-selected area, the edge nodes in the to-be-selected area are sequentially sequenced, a node sequencing table is generated, the request content is cached on the edge node with the first sequencing in the node sequencing table, and when the request content cache is distributed, the scheduling method evaluates the edge nodes, so that the node resource waste is avoided, and the stability of the request content cache is guaranteed;
the invention comprehensively calculates the node load deviation rate through the edge computing equipment at the candidate areaObtaining the sorting coefficient after the discrete degree of the storage capacity, the content estimated capacity index and the user flow floating coefficient at the sampling time pointThe edge calculation at the candidate area is used for comprehensively calculating the multi-source data, so that the calculation efficiency is high, the data processing efficiency is high, and the analysis efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings may be obtained according to these drawings for a person having ordinary skill in the art.
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: referring to fig. 1, the scheduling method of the content distribution network based on edge computation according to the present embodiment includes the following steps:
the monitoring end monitors performance data of all edge nodes in real time, judges whether the edge nodes are abnormal after analyzing the performance data of the edge nodes, and the processing end marks the abnormal edge nodes into an isolation area and stops using the abnormal edge nodes, and sends isolation area information to a network administrator;
the processing end marks the edge node with abnormality into the isolation area and stops using, and the isolation area information is sent to the network manager, which comprises the following steps:
upon determining that an edge node is abnormal, an isolation measure needs to be triggered, which may include ceasing to route new requests to the node, and ceasing to provide service from the node;
generating an isolation zone report containing detailed information about the abnormal node, which may include information on the node's identifier, location, description of performance problems, reason for triggering isolation, etc.;
sending the generated quarantine information to a network administrator or related security team, typically through a secure communication channel, to ensure confidentiality and integrity of the information;
recording the isolation measures performed, including when triggered, why triggered, and how to perform, these recordings can be used as subsequent audit and troubleshooting references;
if the isolation of the edge nodes affects the user experience, it may be necessary to send notifications to the user to interpret the problem and provide backup services or guidance.
The network administrator needs to re-detect and maintain the edge nodes inside the isolation zone, including the following steps:
firstly, the network administrator needs to stop routing new requests to edge nodes within the isolation zone to ensure that new traffic is not continued to be accepted;
performing detailed performance and security detection on edge nodes within the isolation zone to determine existing problems and anomalies, which may include network connection problems, malicious activity, hardware failures, etc.;
based on the detection results, appropriate measures are taken to solve the problem, which may include repairing network connections, removing malware, replacing hardware, etc.;
before the edge nodes in the isolation area are re-started, performance testing and load testing are carried out, so that normal operation of the edge nodes and performance requirements are ensured;
security scrutiny is performed on edge nodes to ensure that they are free of potential vulnerabilities or security issues, and measures can be taken to enhance security;
recording the re-detection and maintenance process, including problems, repair steps, performance test results, etc., which documents can be used for auditing and future reference;
once the edge nodes are identified as repaired and meet the requirements, routing requests to those nodes can resume and re-enter normal operation.
Capturing and analyzing user requests, including information such as content types, user positions, request time and the like, and dividing all edge nodes with the distance from the user being smaller than a distance threshold into a candidate area;
capturing and analyzing user requests, including information of the type of content requested, the user location, the time of the request, etc., and scribing all edge nodes having a distance to the user less than a distance threshold into a candidate area, comprising the steps of:
firstly, the request data of the user needs to be captured and recorded, which can be accomplished by a network device, a proxy server, a log recording system and the like;
analyzing the captured request data, extracting information about the request, which may include content type (e.g., text, image, video), geographic location of the user, request timestamp, etc.;
calculating a distance between each edge node and the user using the geographical location information of the user and the geographical location information of the edge node, the distance calculation may be performed using a Geographical Information System (GIS) tool;
defining a distance threshold, and screening all edge nodes which are smaller than the threshold from the user according to the calculated distance, wherein the nodes are considered as potential candidate areas;
identifying edge nodes meeting the distance condition as candidate areas, a list of candidate nodes may be created or information of the nodes may be recorded.
And after acquiring multiple items of data of the request content and comprehensively analyzing the performance data of the edge nodes, the multiple items of data of the request content and the user data by the edge computing equipment in the area to be selected, sequentially ordering the edge nodes in the area to be selected, generating a node ordering table, and caching the request content to the first edge node in the node ordering table.
Dynamically adjusting policies according to real-time network conditions and user demands of edge nodes, namely automatically reselecting other edge nodes in a candidate area under the condition of high traffic or fault of the current edge node, comprising the following steps:
the request patterns and demands of the users are continuously analyzed to see which content is popular and which nodes are more requested, which helps to determine when reselection is needed;
defining a set of trigger conditions to determine when an automatic reselection of an edge node is required, which may be a combination of conditions of reduced performance, high traffic load, node failure, etc.;
implementing an automatic reselection strategy, automatically selecting one or more standby edge nodes from a to-be-selected area to take over traffic according to a triggering condition, wherein the strategy can select the nodes based on factors such as distance, performance, load balancing and the like;
after selecting the standby node, ensuring smooth handoff of traffic to the new node, which may involve data transfer, cache flushing, and DNS updating;
monitoring the performance of the new nodes to ensure that they can meet the user requirements, and if the new nodes also encounter problems, continuing to reselect;
the reselection and switchover process, including trigger conditions, selected standby nodes, switchover time, etc., are recorded, which aids in troubleshooting and auditing.
Edge computing is a distributed computing model that aims to push computing resources and data processing capabilities to edge locations near data sources and end users, rather than being centralized in a remote data center or cloud platform. Edge computing emphasizes deploying computing, storage, and network resources at the edges of the physical world to provide lower latency, higher bandwidth utilization, and higher reliability, specifically:
proximity data source and end user: edge computing deploys computing resources and services closer to data sources (e.g., sensors, devices) and end users to reduce latency of data transmission;
reducing delay: by performing calculations on the edge devices, the time delay of data transmission from the remote data center to the user device can be reduced; this is very important for real-time applications and applications that are very sensitive to delay;
privacy and security are improved: the data is processed at the edge, so that the requirement of transmitting sensitive data to the cloud can be reduced, and the privacy and safety of the data are improved;
support internet of things (IoT): edge computation is very important for processing data generated by large-scale IoT devices; the method can execute local calculation on the Internet of things equipment, so that dependence on cloud services is reduced;
diversified application scenarios: edge computing is widely used in a variety of fields including smart cities, industrial automation, intelligent transportation, healthcare, video monitoring, virtual reality, and the like;
edge node: edge computing networks typically include edge nodes, which may be physical servers, network devices, or edge devices, distributed at edge locations; they perform local computing tasks in response to the needs of a particular application;
cloud and edge co-operation: typically, edge computing works in conjunction with cloud computing; the edge node can offload part of the computing tasks to the cloud to realize larger-scale data analysis and resource pools;
dynamics and adaptivity: the edge calculation needs to have dynamics so as to adapt to the change of network load and demand; this can be achieved through intelligent scheduling and resource management.
Therefore, in the application, the edge computing equipment is arranged at each edge node, so that the data processing is decentralised, and the processing efficiency is higher.
According to the method, after the performance data of all edge nodes are monitored in real time through the monitoring end, whether the edge nodes are abnormal or not is judged after the performance data of the edge nodes are analyzed, the processing end marks the abnormal edge nodes into the isolation area and stops using the abnormal edge nodes, so that the safety and stability of the request content cache are guaranteed, user requests are captured and analyzed, all the edge nodes with the distance from the user being smaller than the distance threshold value are marked into the area to be selected, multiple data of the request content are obtained, after the performance data of the edge nodes, multiple data of the request content and the user data are comprehensively analyzed through edge computing equipment at the area to be selected, the edge nodes in the area to be selected are sequentially ordered, a node ordering table is generated, the request content is cached on the edge node with the first ordering in the node ordering table, and when the request content cache allocation is carried out, the edge nodes are evaluated, so that the node resource waste is avoided, and the stability of the request content cache is guaranteed.
Example 2: the monitoring end monitors the performance data of all the edge nodes in real time, and judges whether the edge nodes are abnormal or not after analyzing the performance data of the edge nodes;
the performance data comprises node load deviation rate and the discrete degree of storage capacity at the sampling time point;
the computational expression of the node load deviation rate is:wherein->For the node load deviation rate, < >>For the rated load of the edge node, +.>For the real-time load of the edge node, the larger the node load deviation rate is, the more the real-time monitored load of the edge node deviates from the rated load of the edge node, the more the edge node is easy to be abnormal, and the following problems are caused:
1) Node overload: the node with larger load deviation rate can bear too much flow and request, so that the node is overloaded, and the overloaded node can not respond to the request in time, so that the user experience is reduced;
2) The resources are insufficient: overloaded nodes may face problems with insufficient resources, such as processing power, memory, or bandwidth, which may lead to reduced node performance, or even downtime;
3) Network congestion: in some cases, the overloaded node may cause network congestion, affecting the communication quality between other nodes and users;
4) Uneven cache distribution: if the node loads are not balanced, the content distribution of the caches may also be uneven, caches on some nodes may be frequently used, and caches on other nodes may be rarely hit;
5) Usability problem: node load imbalance may result in reduced availability of certain nodes, which may affect whether users can access their desired content;
therefore, when the node load deviation rate is greater than the deviation threshold, the edge node is analyzed for anomalies.
The calculation expression of the discrete degree of the storage capacity at the sampling time point is as follows:in the following，/>Represents the number of sampling time points, +.>Is a positive integer>Representing the remaining storage capacity of the edge node acquired at the ith sampling time point, +.>Representing the average value of the residual storage capacity of the edge node, wherein the greater the dispersion degree of the storage capacity at the sampling time point is, the indicating the edge nodeThe fluctuation of the residual storage capacity in the monitoring time period is large, which indicates that the failure of the storage device of the edge node or the frequent attack of the edge node can cause the following problems:
1) Storage device failure: failure of a storage device can result in a sudden decrease in storage capacity, thereby causing fluctuations;
2) Frequent write and delete operations: if there are a large number of data writing and deleting operations, the storage capacity may fluctuate rapidly; this may be the result of normal operation, but may also be an indication of storage problems;
3) Storing overflow: if the storage device is full, but the data continues to be written, rapid fluctuation of storage capacity may be caused;
4) Storage resource competition: when multiple applications or services compete for resources on the same storage device, fluctuations in storage capacity may result;
5) Malicious activity: malware or attacks may cause abnormal write or delete operations on a storage device to destroy data or raise storage capacity issues;
6) Data leakage: revealing sensitive data may lead to fluctuations in storage capacity, as the data may be deleted or accessed by external malicious subjects;
therefore, when the degree of dispersion of the storage capacity at the sampling time point is larger than the dispersion threshold value, the analysis edge node is abnormal.
Acquiring multiple items of data of the request content, comprehensively analyzing the performance data of the edge nodes, the multiple items of data of the request content and the user data through edge computing equipment at the area to be selected, sequentially ordering the edge nodes in the area to be selected, generating a node ordering table, and caching the request content to the first edge node in the node ordering table;
wherein:
the performance data comprise node load deviation rate and storage capacity discrete degree at sampling time points, the plurality of pieces of data of the requested content comprise content estimated capacity indexes, and the user data comprise user flow floating coefficients;
integration by edge computing devices at candidate areasCalculating node load deviation rate, sampling time point storage capacity discrete degree, content estimated capacity index and user flow floating coefficient to obtain sequencing coefficientThe computational expression is:wherein->Estimating a capacity index for content->Floating coefficient for user traffic, +.>For the node load deviation rate, < >>For the degree of dispersion of the storage capacity at the sampling time point +.>、/>、/>、/>The content estimated capacity index, the user flow floating coefficient, the node load deviation rate and the proportional coefficient of the storage capacity discrete degree at the sampling time point are respectively, and +.>、/>、/>、/>Are all greater than 0;
obtaining ranking coefficientsAfter the value, all edge nodes in the candidate area are added according to the sorting coefficient>Sorting the values from large to small, generating a node sorting table, and caching the request content to the first edge node in the node sorting table;
according to the method, after node load deviation rate, sampling time point storage capacity discrete degree, content estimated capacity index and user flow floating coefficient are comprehensively calculated through edge computing equipment at a candidate area, a sequencing coefficient is obtainedThe edge calculation at the candidate area is used for comprehensively calculating the multi-source data, so that the calculation efficiency is high, the data processing efficiency is high, and the analysis efficiency is improved.
In this application:
content predictive capacity indexThe calculated expression of (2) is: />Wherein->Memory occupied by the content itself>Memory occupied for content history average access +.>Rated capacity for the edge node;
the acquisition logic of the memory occupied by the content history average access is as follows:
acquiring the memory occupied by user request nodes in different time periods when the content is released from the last edge node;
after acquiring the memory occupied by the user request nodes in all time periods, calculating an average value to obtain the memory occupied by the content history average access;
when content is released and accessed by users on the edge node, the more users accessing the content in the same time period, the larger the memory occupation of the edge node is, and therefore, the content estimated capacity index is estimatedThe larger the value, the easier the capacity of the edge node to publish and access the content;
indicating that the edge node requires more memory to store and provide the content may mean that the content is very popular for a period of time or that there is a large number of simultaneous accesses by users, a higher estimated capacity index value may require more memory capacity to meet the user's needs to ensure a quick response to the request while reducing performance problems of the edge node due to insufficient memory.
User flow coefficient of floatThe calculated expression of (2) is: />，/>Requesting real-time user traffic change for edge node content, < >>A period of user login is requested for the edge node content,requesting a period of user logout for the edge node content;
user flow coefficient of floatThe larger the value, the more the number of user accesses the edge node requests content, and thus the more the edge node should act as a content distribution node.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (9)
1. The scheduling method of the content distribution network based on the edge calculation is characterized by comprising the following steps of: the scheduling method comprises the following steps:
s1: the monitoring end monitors the performance data of all the edge nodes in real time, and judges whether the edge nodes are abnormal or not after analyzing the performance data of the edge nodes;
s2: the processing end marks the edge node with abnormality into an isolation area and stops using, and isolation area information is sent to a network administrator;
s3: capturing and analyzing a user request, and scribing all edge nodes with the distance from the user being smaller than a distance threshold value into a candidate area;
s4: acquiring multiple items of data of the request content, comprehensively analyzing the performance data of the edge nodes, the multiple items of data of the request content and the user data through edge computing equipment at the area to be selected, and sequentially sequencing the edge nodes in the area to be selected to generate a node sequencing table;
s5: caching the request content to the first edge node in the node sorting table;
s6: and dynamically adjusting the strategy according to the real-time network condition of the edge node and the user demand.
2. The method for scheduling an edge-computing-based content distribution network according to claim 1, wherein: in step S4, acquiring a plurality of items of data of the request content, and comprehensively analyzing the performance data of the edge node, the plurality of items of data of the request content and the user data through the edge computing equipment at the candidate area;
the performance data comprises node load deviation rate and storage capacity discrete degree at sampling time points, the plurality of data of the requested content comprises content estimated capacity index, and the user data comprises user flow floating coefficient.
3. The method for scheduling an edge-computing-based content distribution network according to claim 2, wherein: in step S4, sequentially sorting edge nodes in the to-be-selected area, and generating a node sorting table includes the following steps:
s4.1: comprehensively calculating node load deviation rate, sampling time point storage capacity discrete degree, content estimated capacity index and user flow floating coefficient by using edge computing equipment at a candidate area to obtain a sequencing coefficientThe computational expression is:wherein->Estimating a capacity index for content->Floating coefficient for user traffic, +.>For the node load deviation rate, < >>For the degree of dispersion of the storage capacity at the sampling time point +.>、/>、/>、/>The content estimated capacity index, the user flow floating coefficient, the node load deviation rate and the proportional coefficient of the storage capacity discrete degree at the sampling time point are respectively, and +.>、/>、/>、/>Are all greater than 0;
s4.2: obtaining ranking coefficientsAfter the value, all edge nodes in the candidate area are added according to the sorting coefficient>And sorting the values from big to small to generate a node sorting table.
4. A method of scheduling a content delivery network based on edge computation according to claim 3, wherein: the user flow rate floating coefficientThe calculated expression of (2) is: />，/>Requesting real-time user traffic change for edge node content, < >>Requesting a period of user login for edge node content, < >>Requesting a user log-out period for edge node content.
5. A method of scheduling a content delivery network based on edge computation according to claim 3, wherein: the content predictive capacity indexThe calculated expression of (2) is: />Wherein->Memory occupied by the content itself>Memory occupied for content history average access +.>Is the rated capacity of the edge node.
6. The method for scheduling an edge-computing-based content distribution network according to claim 5, wherein: the logic for acquiring the memory occupied by the content history average access is as follows:
acquiring the memory occupied by user request nodes in different time periods when the content is released from the last edge node;
and after acquiring the occupied memory of the user request nodes in all the time periods, calculating an average value to obtain the memory occupied by the content history average access.
7. A method of scheduling a content delivery network based on edge computation according to claim 3, wherein: the calculation expression of the node load deviation rate is as follows:wherein->For the node load deviation rate, < >>For the rated load of the edge node, +.>Is the real-time load of the edge node.
8. The method for scheduling an edge-computing-based content distribution network according to claim 7, wherein: the sampling time point is storedThe computational expression of the degree of dispersion of the storage capacity is:in the following，/>Represents the number of sampling time points, +.>Is a positive integer>Representing the remaining storage capacity of the edge node acquired at the ith sampling time point, +.>Representing the average value of the remaining storage capacity of the edge node.
9. The method for scheduling an edge-computing-based content distribution network according to claim 8, wherein: and when the node load deviation rate is larger than a deviation threshold value or the storage capacity discrete degree of the sampling time point is larger than a discrete threshold value, analyzing that the edge node is abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311778147.9A CN117459587B (en) | 2023-12-22 | 2023-12-22 | Scheduling method of content distribution network based on edge calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311778147.9A CN117459587B (en) | 2023-12-22 | 2023-12-22 | Scheduling method of content distribution network based on edge calculation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117459587A CN117459587A (en) | 2024-01-26 |
| CN117459587B true CN117459587B (en) | 2024-03-01 |
Family
ID=89585868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311778147.9A Active CN117459587B (en) | 2023-12-22 | 2023-12-22 | Scheduling method of content distribution network based on edge calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117459587B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104320487A (en) * | 2014-11-11 | 2015-01-28 | 网宿科技股份有限公司 | HTTP dispatching system and method for content delivery network |
| CN110233868A (en) * | 2019-04-20 | 2019-09-13 | 北京工业大学 | A kind of edge calculations data safety and method for secret protection based on Fabric |
| CN113347275A (en) * | 2021-07-06 | 2021-09-03 | 北京云端智度科技有限公司 | Edge node scheduling method and system based on geographic coordinates of user terminal |
| CN114466018A (en) * | 2022-03-22 | 2022-05-10 | 北京有竹居网络技术有限公司 | Scheduling method and device for content distribution network, storage medium and electronic equipment |
| CN116249108A (en) * | 2023-03-01 | 2023-06-09 | 天津理工大学 | Edge computing key management method for trusted uplink of IoT user perception data |
-
2023
- 2023-12-22 CN CN202311778147.9A patent/CN117459587B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104320487A (en) * | 2014-11-11 | 2015-01-28 | 网宿科技股份有限公司 | HTTP dispatching system and method for content delivery network |
| CN110233868A (en) * | 2019-04-20 | 2019-09-13 | 北京工业大学 | A kind of edge calculations data safety and method for secret protection based on Fabric |
| CN113347275A (en) * | 2021-07-06 | 2021-09-03 | 北京云端智度科技有限公司 | Edge node scheduling method and system based on geographic coordinates of user terminal |
| CN114466018A (en) * | 2022-03-22 | 2022-05-10 | 北京有竹居网络技术有限公司 | Scheduling method and device for content distribution network, storage medium and electronic equipment |
| CN116249108A (en) * | 2023-03-01 | 2023-06-09 | 天津理工大学 | Edge computing key management method for trusted uplink of IoT user perception data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117459587A (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018161447A1 (en) | Protection method and system for cdn client source station | |
| US10048996B1 (en) | Predicting infrastructure failures in a data center for hosted service mitigation actions | |
| US20100037225A1 (en) | Workload routing based on greenness conditions | |
| CN110213203B (en) | Network scheduling method and device and computer storage medium | |
| US11693563B2 (en) | Automated tuning of a quality of service setting for a distributed storage system based on internal monitoring | |
| US11637902B2 (en) | Curating proxy server pools | |
| WO2020134364A1 (en) | Virtual machine migration method, cloud computing management platform, and storage medium | |
| US11392315B1 (en) | Automatically tuning a quality of service setting for a distributed storage system with a deep reinforcement learning agent | |
| CN105847377A (en) | Cluster network's request congestion and overload processing method and system | |
| CN108989368A (en) | A kind of control method and monitoring device of link-quality | |
| Mogylevych et al. | Improved estimates for the reliability indicators of information and communication network objects with limited source information | |
| CN116418653A (en) | Fault positioning method and device based on multi-index root cause positioning algorithm | |
| US20090150564A1 (en) | Per-user bandwidth availability | |
| CN117459587B (en) | Scheduling method of content distribution network based on edge calculation | |
| CN109120424B (en) | Bandwidth scheduling method and device | |
| CN114362994A (en) | Multilayer different-granularity intelligent aggregation railway system operation behavior safety risk identification method | |
| WO2024088025A1 (en) | Automated 5gc network element management method and apparatus based on multi-dimensional data | |
| CN104135525A (en) | Resource expanding method and device for cloud platform ELB components | |
| CN115277080B (en) | Content distribution network cache pollution defense method based on merck tree | |
| CN110445803A (en) | A kind of traffic smoothing moving method of isomery cloud platform | |
| CN114124797B (en) | Server routing method and device, electronic equipment and storage medium | |
| CN116915841A (en) | CDN edge node hot spot spreading method, device, equipment and storage medium | |
| CN114493339A (en) | Power grid information safety early warning system based on data feature extraction | |
| CN116962395A (en) | Cloud node determining method, device, equipment and computer readable storage medium | |
| Wang et al. | Lightweight IT operation and maintenance integrated monitoring method for APP system |
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 |