WO2022035391A1 - Server assignment by traffic modelling in content delivery networks (cdn) - Google Patents
Server assignment by traffic modelling in content delivery networks (cdn) Download PDFInfo
- Publication number
- WO2022035391A1 WO2022035391A1 PCT/TR2020/051310 TR2020051310W WO2022035391A1 WO 2022035391 A1 WO2022035391 A1 WO 2022035391A1 TR 2020051310 W TR2020051310 W TR 2020051310W WO 2022035391 A1 WO2022035391 A1 WO 2022035391A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- pop
- module
- bandwidth
- traffic
- Prior art date
Links
- 238000007781 pre-processing Methods 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 238000013480 data collection Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000002708 enhancing effect Effects 0.000 claims abstract 3
- 238000013528 artificial neural network Methods 0.000 claims description 22
- 238000000034 method Methods 0.000 claims description 18
- 230000005540 biological transmission Effects 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 10
- 238000007619 statistical method Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001373 regressive effect Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram 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
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L61/00—Network arrangements, protocols or services for addressing or naming
- H04L61/45—Network directories; Name-to-address mapping
- H04L61/4505—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols
- H04L61/4511—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols using domain name system [DNS]
Abstract
A performance enhancing system in Content Delivery Networks (CDN) characterised in that; including, - Data Collection Module (100) that collects data from the PoPs of the CDN, - Data Pre-Processing Module (200) used to make data available for use, - Traffic Modelling Module (300), which creates traffic modelling by self-similar analysis in CDNs and - PoP Assignment List Module (400) that creates active PoP Assignment Lists (402) by processing the results of the Traffic Modelling Module (300) and data from other modules (100, 200).
Description
Server Assignment by Traffic Modelling in Content Delivery Networks (CDN)
Technical Field
The invention relates to a method and system to increase performance in Content Delivery Networks (CDN).
Prior Art
In the prior art, content requests from the end user reach our DNS servers located in various parts of the world as Anycast and are directed to the relevant PoP according to the predetermined DNS rules by considering the location information of the end user.
Fixed routing tables prepared according to these rules do not make the daily / monthly / annual changes in internet traffic and statistical analysis of internet traffic while directing end user requests to PoPs. Due to such static PoP lists, content transmission speed decreases and performance loss occurs. In addition, fixed PoP routing tables made by looking at the user location cause the content not to be transmitted over the PoP, which is more available at that time. Therefore, static PoP routing tables cause users to reach content later.
On the other hand; since the static assignment list in the PoPs redirects without considering the internet bandwidth, the internet bandwidth contracts of our company are made according to the traffic with the highest internet. Therefore, fixed PoP assignment lists increase the internet lease contract costs of bandwidth PoPs.
Purpose of the Invention
Content Delivery Networks (CDN) aim to deliver the contents, such as images, videos, etc., quickly to users in different parts of the world by caching them in Points of Presence (PoP) in different geographical locations.
Active PoP assignment lists were created by modelling the network data with Auto- Regressive Integrated Moving Average and Artificial Neural Networks according to their statistical self-similar characteristics, and by providing the outputs as input to Artificial Intelligence. Although self-similar analysis is a generally accepted phenomenon in the network literature, a PoP assignment list was not created by performing traffic modelling by self-similar analysis in Content Distribution Networks (CDN). Traffic modelling in Content Delivery Networks (CDN) plays a key role in network operations and management. The purpose of the invention is, by modelling Content Delivery Network (CDN) traffic, to ensure predictability of the traffic volume and PoP assignments are made according to this data.
The purpose of the invention is to ensure the analysis of daily/monthly/annual changes in internet traffic, while fixed routing tables and end user requests are forwarded to PoPs, and statistical analysis of internet traffic.
Another purpose of the invention is to ensure that millions of user requests are directed according to active PoP assignment lists to be created using internet data.
Another purpose of the invention is to reduce internet contract costs by routing according to active PoP routing lists to PoPs that are more available in terms of traffic bandwidth on the Content Delivery Network (CDN).
The system developed to achieve the mentioned objectives includes the Data Acquisition Module (100), the Data Pre-Processing Module (200), the Traffic Modelling Module (300) and the PoP Assignment List Module (400).
Description of Figures
The attached Figure - 1 is the schematic view of a CDN built on the existing internet network.
Figure-2 is the data acquisition and work-flow diagram of the main module.
Figure-3 is the detailed work-flow chart of the system subject to the invention.
The main elements expressed in the figures are given below with their reference numbers and names.
(100) Data Acquisition Module
(101) Bandwidth raw data
(102) PoP Raw Data
(103) DNS Raw Data
(104) PoP measurements
(200) Data Pre-Processing Module
(201) Bandwidth Data
(202) PoP Data
(203) DNS Data
(300) Traffic Modelling Module
(301) Bandwidth Data
(302) Self-Similarity Analysis
(303) Training Data
(304) Test Data
(305) S AR. IM A
(306) Artificial Neural Network 1
(307) Model Parameters
(308) Optimal Model
(309) Conclusion
(400) PoP Assignment List Module
(401) Artificial Neural Network 2
(402) PoP Assignment List
Explanation of the Invention
The invention is a performance improving product in Content Delivery Networks (CDN).
Content Delivery Networks (CDN) aim to deliver the contents, such as images, videos, etc., quickly to users in different parts of the world by caching them in Points of Presence (PoP) in different geographical locations. CDNs consist of many PoPs built on the existing global internet network. PoPs play an important role in content delivery: With well-designed PoPs, content transmission times are reduced and end users access content faster.
Content requests from the end user reach our DNS servers located in various parts of the world by Anycast method. Anycast method is that the same IP (Internet Protocol) address is valid for many servers, thus, the DNS query of the end user is directed to the nearest DNS server thanks to Anycast. After the DNS server, the end user request is directed to the relevant PoP according to the predetermined DNS rules, considering the location information of the end user.
In the systems in the prior art, depending on the location of the end user, the PoP to be directed to is determined according to the determined and unchanging PoP assignment order. End user requests are routed by an algorithm that is not updated and has different weights given to different PoPs in the existing CDN systems.
With the invention, millions of user requests are directed according to active PoP assignment lists to be created using internet data. Internet contract costs are reduced by directing the traffic on the CDN to PoPs that are more available in terms of bandwidth according to active PoP routing lists.
Active PoP assignment lists were created by modelling the network data with Auto- Regressive Integrated Moving Average and Artificial Neural Networks according
to their statistical self-similar characteristics, and by providing the outputs as input to Artificial Intelligence. Although self-similar analysis is a generally accepted phenomenon in the network literature, a PoP assignment list was not created by performing traffic modelling by self-similar analysis in Content Distribution Networks (CDN). Traffic modelling in Content Delivery Networks (CDN) plays a key role in network operations and management. Via the invention, traffic volume is estimated by modelling Content Delivery Network (CDN) traffic and PoP assignments are made according to this data.
In Figure 1, a CDN built on the existing internet network is schemed. CDN PoPs in clusters of different Internet Service Providers (ISP) constitute the Content Delivery Network. Data Collection Module (100), one of the building blocks of the system subject to the invention, is located in each PoP and provides data flow to the Main Module.
There are four main modules in the system which is the subject of the invention: Data Collection Module (100), which collects data from the PoPs of the CDN, transmits the data to the Data Pre-Processing Module (200) in order to make the data available for use. Then the outputs of the Data Pre-Processing Module (200) are transmitted to the Traffic Modelling Module (300). Active PoP Assignment Lists (402) are created by transmitting the results of the Traffic Modelling Module (300) and the data from other modules (100, 200) to the PoP Assignment List Module (400). Finally, the PoP Assignment List (402) made for each PoP is sent to its own PoP and begins to be used for end user routing.
In the four main modules (100, 200, 300, 400) in the system of the invention, the algorithm transmits the PoP lists to PoPs to be used in the future after processing the past time zone data for a certain period of time. The current data collected in the period when the current PoP lists are valid, re-creates the said algorithm and provides the data input by creating the PoP assignment lists (402) in another time to come. The Data Collection Module (100) will send data to the modules ahead at certain intervals and PoP Assignment Lists (402) valid for a certain time interval will be made. Since the PoP Assignment Lists (402) will be updated at certain time
intervals, dynamic PoP Assignment Lists, which are the opposite of the fixed PoP Assignment Lists (402), are taken into use on the CDN.
Bandwidth Raw Data (101), PoP Raw Data (102), DNS Raw Data (103) are taken from CDN PoPs and RAM, CPU, Disk VO Measurements are taken from PoPs (104); and are collected in the Data Collection Module (100). The collected raw data (101, 102, 103) are transmitted to the Data Pre-Processing Module (200) for processing. PoP Measurements (104) collected from PoPs are directly transmitted to the PoP Assignment List Module (400).
Raw data (101, 102, 103) are processed in the Data Pre-Processing Module (200). Processed Bandwidth Data (201) is transmitted to the Traffic Modelling Module (300). Processed PoP Data (202) and DNS Data (203) are transmitted as input to the PoP Assignment List Module (400).
In the Traffic Modelling Module (300), the processed Bandwidth Data (201) coming from the Data Pre-Processing Module (200) is divided into time segments and started to be used as Bandwidth Data (301) in the Traffic Modelling Module (300). The Partial Bandwidth Data (301) is subjected to Self-Similarity analysis (302) for statistical analysis. After the said analysis, the piecewise bandwidth Data is divided into two parts and Training Data (303) and Test Data (304) are obtained. Training Data (303) is analysed according to the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) (305) method and by building the Artificial Neural Network 1 (306). At the end of the mentioned analyses, the most suitable Model Parameters (307) of the Training Data (303) are found. After determining the parameters of the models, using the Test Data (304), the Optimal Model (308) is found by error analysis. The most optimal model (308), after found as a result of error analysis, is sent as the output of the Traffic Modelling Module (300) to the PoP Assignment List Module (400) to create its input.
In the PoP Assignment List Module (400); PoP measurements (104) from Data Collection Module (100), PoP Data (202) and DNS Data (203) from Data Pre-
Processing Module (200) and Result (309) from Traffic Modelling Module (300) enters Artificial Neural Network 2 (401) to generate the PoP Assignment List (402). At the end of the mentioned process, a PoP Assignment List (402) with a specific time interval criterion is created specifically for each PoP. The kinetic PoP Assignment List (402) is put into use by loading PoP Assignment Lists to the PoPs in the CDN and updating them at certain time intervals.
The processes carried out by the system subject to the invention are as follows;
Transmission of Bandwidth Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
Transmission of PoP Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
Transmission of DNS Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
Transmission of PoP Measurements (104) (CPU/RAM/IO) from Data Collection Module (100) to PoP Assignment List Module (400), Transmitting the Bandwidth Data (200) ready to use in the Data Pre- Processing Module (201) to the Traffic Modelling Module (300), Transmitting the ready-to-use PoP Data in the Data Pre-Processing Module (200) to (202) PoP Assignment List Module (400),
Transmission of the DNS Data (203) ready to use in the Data Pre-Processing Module (200) to the PoP Assignment List Module (400),
- Dividing Bandwidth Data (301) into appropriate time intervals and transmitting them for Self-similarity Analysis (302),
- Keeping a part of the Self-Similar Analysis (302) data as Training Data (303),
- Keeping a part of the Self-Similar Analysis (303) data as Test Data (302),
- Modelling of Training Data by constructing (303) SARIMA (305) method and Artificial Neural Network 1 (306),
Determination of Model Parameters (307) of SARIMA (305) method and Artificial Neural Network 1 (306) results,
Transmission of Model Parameters (307) to the Optimal Model (308) for calculating the margin of error of the traffic model,
Transmission of Bandwidth Test Data (304) to the Optimal Model (308) for calculation of margins of error,
Transmission of the most pleasing error, thus the most valid Traffic Modelling Result (309) to Artificial Neural Network 2 (401) in the PoP Assignment List Module (400),
Sending the results of Artificial Neural Network 2 (401) to the PoP Assignment List (402) to form the PoP assignment lists.
The bandwidth raw data (101) included in the invention refers to the current Bandwidth data from PoPs.
PoP Raw Data (102) refers to the current PoP Data from PoPs.
DNS Raw Data (103) refers to the current DNS Data from PoPs.
CPU, RAM, Disk Input / Output data from PoPs are collected at PoP Measurements (104) point.
Bandwidth Data (201) refers to the processed form of Bandwidth Raw Data (101). Bandwidth Data (201) is an input to the Traffic Modelling Module (300).
PoP Raw data (102) is processed and creates PoP Data (202) as input to the PoP Assignment Module (300).
DNS Raw data (103) is processed and creates DNS Data (203) as input to PoP Assignment Module (300).
Bandwidth Data (301) received from Data Pre-Processing Module (200) are divided into appropriate time intervals.
Statistical analysis of Bandwidth Data (301) divided into time segments is performed with Self-Similarity Analysis (302) method.
A part of the available Bandwidth data (301) is reserved as a Training Data (303) and a part as a Test Data (304).
Traffic modelling is performed with the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) (305) method using Bandwidth Training Data (303). In addition, traffic modelling is also performed by building Artificial Neural Network 1 (306) using the mentioned Bandwidth Training Data (303). Using the modelling, SARIMA (305) and the least error-producing model parameters (307) of the Artificial Neural Network are determined.
Using the Model Parameters and performing Error Analysis with Bandwidth Test Data, it is determined whether the best method in traffic modelling will be SARIMA (305) or Artificial Neural Network-1 (306). The mentioned process is done by error analysis. As a result, the Optimal Model (308) is determined.
The best traffic modelling results (309) are calculated and sent as input to the PoP Assignment List Module (400).
Artificial Neural Network 2 (401) is created by using PoP measurements (RAM / CPU / IO) from Data Collection Module (100) and PoP Data (202) from Data Pre- Processing Module (200) + DNS Data (203) and the results from Traffic Modelling Module (300). As a result of Artificial Neural Network 2 (401) calculations, PoP Assignment Lists (402) are determined for each PoP.
Claims
C L A I M S A performance enhancing system in Content Delivery Networks (CDN) characterised in that; including,
- Data Collection Module (100) that collects data from the PoPs of the CDN,
- Data Pre-Processing Module (200) used to make data available for use,
- Traffic Modelling Module (300), which creates traffic modelling by self-similar analysis in CDNs and
- PoP Assignment List Module (400) that creates active PoP Assignment Lists (402) by processing the results of the Traffic Modelling Module (300) and data from other modules (100, 200). The Data Collection Module (100) mentioned in claim 1 characterised in that; including,
- Bandwidth Raw Data (101) containing current Bandwidth Data from PoPs,
- PoP Raw Data (102) containing up-to-date PoP Data from PoPs,
- DNS Raw Data (103), which includes current DNS Data from PoPs,
- PoP Measurements (104) in which CPU, RAM, Disk Input/Output data from PoPs are collected. The Data Pre-Processing Module (200) mentioned in claim 1 characterised in that; including,
- Bandwidth Data (201), which is input to Traffic Modelling Module (300) by processing Bandwidth Raw Data (101),
- PoP data (202), which is input to the PoP Assignment List Module (400) by processing the PoP Raw data (102),
- DNS Data (203), which is input to the PoP Assignment List Module (400) by processing the DNS Raw data (103).
The Traffic Modelling Module (300) mentioned in claim 1 characterised in that; including,
- Bandwidth Data (301) dividing the Bandwidth Data (201) received from the Data Pre-Processing Module (200) into appropriate time intervals,
- Self-Similarity Analysis (302), which performs statistical analysis of Bandwidth Data divided into time segments (301), with Self-similarity method,
- Training Data (303) separated from the current Bandwidth (301) data,
- Test Data (304) separated from Current Bandwidth (301) data,
- SARIMA (305), which performs traffic modelling with the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) method using Training Data (303),
- Artificial Neural Network 1 (306), which performs traffic modelling by using Training Data (303) and building an Artificial Neural Network,
- Model Parameters (307) determining the least error model parameters of SARIMA (305) and Artificial Neural Network 1 (306),
- The Optimal Model (308) that determines whether the best method in traffic modelling will be SARIMA (305) or Artificial Neural Network- 1 by using Model Parameters (307) and Error Analysis with Bandwidth Test Data (304),
- Result (309) that calculates the best traffic modelling outputs and sends it to the PoP Assignment List Module (400) as input. The PoP Assignment List Module (400) mentioned in Claim 1 characterised in that; including,
- Artificial Neural Network-2 (401), which creates the PoP Assignment List (402) by processing the results (309) from (300) Traffic Modelling Module with PoP measurements (101) (RAM / CPU / IO) from Data
Collection Module (100) and PoP Data (202) + DNS Data (203) from Data Pre-Processing Module (200),
- PoP Assignment List (402) that determines the PoP Assignment Lists (402) for each PoP as a result of Artificial Neural Network - 2 (401) calculations The invention is a performance enhancing method in Content Delivery Networks (CDN) and is characterised in that; including the steps of
- Transmission of Bandwidth Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
- Transmission of PoP Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
- Transmission of DNS Raw Data (101) from Data Collection Module (100) to Data Pre-Processing Module (201),
- Transmission of PoP Measurements (104) (CPU / RAM / IO) from Data Collection Module (100) to PoP Assignment List Module (400),
- Transmitting the Bandwidth Data (200) ready to use in the Data Pre- Processing Module (201) to the Traffic Modelling Module (300),
- Transmission of the PoP Data (202) ready to use in the Data Pre- Processing Module (200) to the PoP Assignment List Module (400),
- Transmission of the DNS Data (203) ready to use in the Data Pre- Processing Module (200) to the PoP Assignment List Module (400),
- Dividing Bandwidth Data (301) into appropriate time intervals and transmitting them for Self-similarity Analysis (302),
- Keeping a part of the Self-Similar Analysis (302) data as Education Data (303),
- Keeping a part of the Self-Similar Analysis (303) data as Test Data (302),
- Modelling Training Data (303) by constructing the SARIMA method (305) and Artificial Neural Network 1 (306),
- Determination of Model Parameters (307) of the results of the SARIMA method (305) and Artificial Neural Network 1 (306),
- Transmitting the traffic model whose Model Parameters (307) are determined, to the Optimal Model for calculating the margin of error,
- Transmitting Bandwidth Test Data (304) to the Optimal Model (308) for calculating margins of error, - Transmitting the most accurate, most valid Traffic Modelling Result
(309) to Artificial Neural Network 2 (401) in the PoP Assignment List Module (400),
- Sending the results of Artificial Neural Network 2 (401) to the PoP Assignment List (402) to create the PoP assignment lists.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TR2020/12584A TR202012584A2 (en) | 2020-08-11 | 2020-08-11 | Server Assignment with Traffic Modeling in Content Distribution Networks (CDN) |
| TR2020/12584 | 2020-08-11 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022035391A1 true WO2022035391A1 (en) | 2022-02-17 |
Family
ID=75526555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/TR2020/051310 WO2022035391A1 (en) | 2020-08-11 | 2020-12-16 | Server assignment by traffic modelling in content delivery networks (cdn) |
Country Status (2)
| Country | Link |
|---|---|
| TR (1) | TR202012584A2 (en) |
| WO (1) | WO2022035391A1 (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140164584A1 (en) * | 2012-12-07 | 2014-06-12 | Verizon Patent And Licensing Inc. | Selecting a content delivery network |
| CN105978733B (en) * | 2016-06-27 | 2019-04-16 | 华北理工大学 | A kind of Modeling Network Traffic method and system based on Wei Buer distribution |
| CN110460591A (en) * | 2019-07-26 | 2019-11-15 | 南京理工大学 | Based on the CDN Traffic anomaly detection device and method for improving separation time memory network |
-
2020
- 2020-08-11 TR TR2020/12584A patent/TR202012584A2/en unknown
- 2020-12-16 WO PCT/TR2020/051310 patent/WO2022035391A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140164584A1 (en) * | 2012-12-07 | 2014-06-12 | Verizon Patent And Licensing Inc. | Selecting a content delivery network |
| CN105978733B (en) * | 2016-06-27 | 2019-04-16 | 华北理工大学 | A kind of Modeling Network Traffic method and system based on Wei Buer distribution |
| CN110460591A (en) * | 2019-07-26 | 2019-11-15 | 南京理工大学 | Based on the CDN Traffic anomaly detection device and method for improving separation time memory network |
Non-Patent Citations (1)
| Title |
|---|
| IEEE SERVICE CENTER, vol. 1, no. 1, 1 January 2001 (2001-01-01), PISCATAWAY, NJ, US, pages 94 - 95, XP011394446, ISSN: 1053-5888, DOI: 10.1109/MSP.2001.911202 * |
Also Published As
| Publication number | Publication date |
|---|---|
| TR202012584A2 (en) | 2020-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | End-user mapping: Next generation request routing for content delivery | |
| CN102792292B (en) | The system and method for site performance optimization and internet service process | |
| CN105162900B (en) | A kind of domain name mapping of multi-node collaboration and caching method and system | |
| US9948709B2 (en) | Using resource timing data for server push in multiple web page transactions | |
| CN103051629A (en) | Software defined network-based data processing system, method and node | |
| CN110430274A (en) | A kind of document down loading method and system based on cloud storage | |
| CN103825975A (en) | Cdn node distribution server and system | |
| US11323412B2 (en) | DNS rendezvous localization | |
| US20190222675A1 (en) | Node route selection method and system | |
| US20200175419A1 (en) | Multi-machine based collaborative learning | |
| CN103685056A (en) | Load balancing method and load balancing system | |
| KR20190012928A (en) | Load balancing apparatus and method | |
| CN103401799A (en) | Method and device for realizing load balance | |
| JP5871908B2 (en) | Method and system for controlling data communication within a network | |
| WO2022035391A1 (en) | Server assignment by traffic modelling in content delivery networks (cdn) | |
| CN112182070A (en) | Elasticissearch query management system based on cloud computing | |
| KR101131787B1 (en) | Method for updating data stored in cache server, cache server and content delivery system thereof | |
| CN103916328A (en) | Flow control method and system for content distribution network and peer-to-peer network | |
| CN114520784B (en) | Dynamic content acceleration access method and device | |
| CN111818590B (en) | Wireless network path optimization method and device | |
| CN112187839B (en) | Distribution method, server and storage medium | |
| Maiti et al. | Node allocation in Peer-to-peer overlay networks based remote instrumentation with smart devices | |
| CN108259337B (en) | ICN routing method and system based on interest network fragmentation | |
| CN108881939B (en) | Video CDN processing method and system | |
| CN107920104A (en) | A kind of method and system of cluster server caching load balancing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20949647 Country of ref document: EP Kind code of ref document: A1 |