Resource scheduling method and device based on service measurement and calculation, medium and product

Abstract

The invention discloses a resource scheduling method and device based on service measurement and calculation, a medium and a product. The method comprises the following steps: acquiring scheduling resource data of flow service scheduling of a local area; carrying out service quality and flow direction analysis on the scheduling resource data, and determining an evaluation index; predicting the bandwidth utilization rate and the service quality index of the next period according to the evaluation index, and scoring the bandwidth utilization rate and the service quality index respectively; determining the distribution weight of the local IDC analysis fast table according to the scoring result; and selecting the local IDC analysis fast table according to the distribution weight to perform scheduling query, and determining a resource scheduling address. According to the scheme, the resource scheduling precision and efficiency can be improved, and the user business service quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a resource scheduling method, device, medium and product based on service measurement. Background Art

[0002] CDN (Content Delivery Network) is a technology widely used in network services. It improves the speed and quality of users' access to websites by distributing content to nodes around the world. In this network, GSLB (Global Server Load Balance) is usually implemented through intelligent DNS (Domain Name Serve), which distributes content to the nearest cache server that can be requested based on the address of the LDNS (Local Domain Name Serve) configured by the customer. Therefore, the scheduling accuracy of CDN needs to rely on the intelligent resolution of GSLB.

[0003] The existing technology obtains the running status data of CDN and performs scheduling based on the availability and service quality of the server's CPU, memory, hard disk, etc. However, the scheduling strategy of the existing technology simply calculates the server availability, and does not perform dynamic scheduling based on the weight of the expected end-to-end service quality indicators and the expected resource load. The resource scheduling accuracy and efficiency are low, and the user service quality is low. Summary of the invention

[0004] Compared with the prior art, the present invention proposes a resource scheduling method, device, medium and product based on business measurement, which can improve the accuracy and efficiency of resource scheduling and improve the user business service quality.

[0005] An embodiment of the present invention provides a resource scheduling method based on business measurement, the method comprising:

[0006] Collect scheduling resource data for traffic service scheduling in this area;

[0007] Analyze the service quality and traffic flow of the scheduling resource data to determine evaluation indicators;

[0008] Predicting the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and scoring them respectively;

[0009] Determine the allocation weight of the local IDC resolution fast table based on the scoring results;

[0010] The local IDC resolution fast table is selected according to the allocation weight for scheduling query to determine the resource scheduling address.

[0011] Preferably, the scheduling resource data for the traffic service scheduling in the region is collected, including:

[0012] Obtain GSLB scheduling data, DNS-related IDC IP resolution data, regional network port export DPI traffic data, and IDC export DPI traffic data for the region's traffic services;

[0013] Synchronously backing up the GSLB scheduling data;

[0014] The acquired data is converted into a format and normalized to synthesize the acquired data into an XDR call record.

[0015] Furthermore, the service quality and traffic flow direction of the scheduling resource data are analyzed to determine evaluation indicators, including:

[0016] Determine the drill-down levels of different IDC customers, different services, and different domain names according to the customer information, IP address library information, and domain name information in the scheduling resource data;

[0017] Determine the service quality indicators by analyzing the specific service quality and domain name quality at different drill-down levels;

[0018] Determine the bandwidth utilization of IDC-related links by analyzing the service traffic flow rates of websites at different drill-down levels;

[0019] The service quality index and the bandwidth utilization rate are used as the evaluation indexes.

[0020] Furthermore, service quality indicators are determined, including:

[0021] Data association is performed between the preset IP address database and the IPs drilled down by SPs of major business categories, and high-quality and low-quality IPs are analyzed. The secondary domain names to which they belong are marked, and the HTTP response success rate, HTTP response delay, TCP establishment success, and TCP establishment delay of the IPs are determined as business indicators.

[0022] The preset domain name library, the IP address library and the secondary domain names drilled down by the SP of the major business category are associated with each other, and high-quality domain names and low-quality domain names are analyzed. The SP and IP to which they belong are marked in the secondary domain name data, and the HTTP response success rate, HTTP response delay, TCP establishment success and TCP establishment delay of the secondary domain name are determined as business indicators.

[0023] As a preferred solution, the bandwidth utilization of IDC related links is determined, including:

[0024] The preset IP address database is associated with the first-level domain name drilled down by the SP of the business category to generate the peak flow rate and load status of each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers;

[0025] The IP address database is associated with the second-level domain name drilled down from the first-level domain name to generate peak flow rates and load conditions of each second-level domain name under each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers;

[0026] Determine the bandwidth utilization of IDC-related links based on traffic analysis.

[0027] Preferably, the bandwidth utilization and service quality index of the next cycle are predicted according to the evaluation index, and are scored respectively, including:

[0028] Calculate the bandwidth utilization and service quality indicators collected during this period to determine the stability value, trend value and seasonality value;

[0029] The triple exponential smoothing method is used to calculate the stability value, trend value and seasonal value of bandwidth utilization and service quality indicators respectively, and predict the bandwidth utilization rate of IP in the next cycle and the related service indicators of IP in the next cycle;

[0030] The calculated bandwidth utilization of the IP in the next cycle and related business indicators of the IP in the next cycle are scored respectively according to the pre-divided indicator scoring intervals.

[0031] Preferably, the allocation weight of the local IDC resolution fast table is determined according to the scoring result, including:

[0032] The bandwidth utilization rate and service quality index scores of the same IP are weighted and summed to determine the weight of each IP of the domain name;

[0033] The IP weights of all IPs under the same domain name are summed up, and the allocation weights of different domain names in the local IDC resolution fast table are updated.

[0034] Preferably, selecting the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address includes:

[0035] Searching according to the local IDC parsing quick table;

[0036] When the resource address is found and the corresponding allocation weight is not 0, it is fed back to the LDNS for resource scheduling;

[0037] When the resource address is not found, or the allocation weight corresponding to the found resource address is 0, the GSLB scheduling data in the scheduling resource data is parsed and searched, and the latest resource address found is fed back to the LDNS for resource scheduling.

[0038] The embodiment of the present invention further provides a resource scheduling device based on business measurement, the device comprising:

[0039] The data collection module is used to collect the scheduling resource data of the traffic service scheduling in this area;

[0040] An analysis module, used to analyze the service quality and traffic flow of the scheduling resource data and determine evaluation indicators;

[0041] A prediction module, used to predict the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and score them respectively;

[0042] The weight module is used to determine the allocation weight of the local IDC resolution fast table according to the scoring results;

[0043] The scheduling module is used to select the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address.

[0044] Preferably, the data acquisition module is specifically used for:

[0045] Obtain GSLB scheduling data, DNS-related IDC IP resolution data, regional network port export DPI traffic data, and IDC export DPI traffic data for the region's traffic services;

[0046] Synchronously backing up the GSLB scheduling data;

[0047] The acquired data is converted into a format and normalized to synthesize the acquired data into an XDR call record.

[0048] Preferably, the analysis module is specifically used for:

[0049] Determine the drill-down levels of different IDC customers, different services, and different domain names according to the customer information, IP address library information, and domain name information in the scheduling resource data;

[0050] Determine the service quality indicators by analyzing the specific service quality and domain name quality at different drill-down levels;

[0051] Determine the bandwidth utilization of IDC-related links by analyzing the service traffic flow rates of websites at different drill-down levels;

[0052] The service quality index and the bandwidth utilization rate are used as the evaluation indexes.

[0053] Furthermore, the analysis module is specifically used for:

[0054] Data association is performed between the preset IP address database and the IPs drilled down by SPs of major business categories, and high-quality and low-quality IPs are analyzed. The secondary domain names to which they belong are marked, and the HTTP response success rate, HTTP response delay, TCP establishment success, and TCP establishment delay of the IPs are determined as business indicators.

[0055] The preset domain name library, the IP address library and the secondary domain names drilled down by the SP of the major business category are associated with each other, and high-quality domain names and low-quality domain names are analyzed. The SP and IP to which they belong are marked in the secondary domain name data, and the HTTP response success rate, HTTP response delay, TCP establishment success and TCP establishment delay of the secondary domain name are determined as business indicators.

[0056] Furthermore, the analysis module is specifically used for:

[0057] The preset IP address database is associated with the first-level domain name drilled down by the SP of the business category to generate the peak flow rate and load status of each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers;

[0058] The IP address database is associated with the second-level domain name drilled down from the first-level domain name to generate peak flow rates and load conditions of each second-level domain name under each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers;

[0059] Determine the bandwidth utilization of IDC-related links based on traffic analysis.

[0060] Preferably, the prediction module is specifically used for:

[0061] Calculate the bandwidth utilization and service quality indicators collected during this period to determine the stability value, trend value and seasonality value;

[0062] The triple exponential smoothing method is used to calculate the stability value, trend value and seasonal value of bandwidth utilization and service quality indicators respectively, and predict the bandwidth utilization rate of IP in the next cycle and the related service indicators of IP in the next cycle;

[0063] The calculated bandwidth utilization of the IP in the next cycle and related business indicators of the IP in the next cycle are scored respectively according to the pre-divided indicator scoring intervals.

[0064] Preferably, the weight module is specifically used for:

[0065] The bandwidth utilization rate and service quality index scores of the same IP are weighted and summed to determine the weight of each IP of the domain name;

[0066] The IP weights of all IPs under the same domain name are summed up, and the allocation weights of different domain names in the local IDC resolution fast table are updated.

[0067] Preferably, the scheduling module is specifically used for:

[0068] Searching according to the local IDC parsing quick table;

[0069] When the resource address is found and the corresponding allocation weight is not 0, it is fed back to the LDNS for resource scheduling;

[0070] When the resource address is not found, or the allocation weight corresponding to the found resource address is 0, the GSLB scheduling data in the scheduling resource data is parsed and searched, and the latest resource address found is fed back to the LDNS for resource scheduling.

[0071] An embodiment of the present invention also provides a resource scheduling device based on business measurement, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a resource scheduling method based on business measurement as described in any of the above embodiments.

[0072] An embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a resource scheduling method based on business measurement as described in any of the above embodiments.

[0073] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0074] Compared with the prior art, the present invention provides a resource scheduling method, device, medium and product based on business measurement, which collects scheduling resource data of traffic business scheduling in this area; analyzes the service quality and traffic flow direction of the scheduling resource data to determine the evaluation index; predicts the bandwidth utilization rate and service quality index of the next cycle according to the evaluation index, and scores them respectively; determines the allocation weight of the local IDC resolution fast table according to the scoring result; selects the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address. The application scheme can improve the accuracy and efficiency of resource scheduling and improve the service quality of user business. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A schematic diagram of a process flow of a resource scheduling method based on business measurement provided by an embodiment of the present invention;

[0076] Figure 2 is another flow chart of a resource scheduling method based on business measurement provided by an embodiment of the present invention;

[0077] Figure 3 It is a schematic diagram of a resource level framework for resource data collection provided by an embodiment of the present invention;

[0078] Figure 4 It is a flow chart of the quality analysis and acquisition process provided by the embodiment of the present invention;

[0079] Figure 5 It is a flow chart of flow direction analysis provided by an embodiment of the present invention;

[0080] Figure 6 is a business flow chart of a resource scheduling method based on business measurement provided by an embodiment of the present invention;

[0081] Figure 7 is a schematic diagram for comparing resource scheduling processes provided by an embodiment of the present invention;

[0082] Figure 8 It is a structural diagram of a resource scheduling device based on business measurement provided by an embodiment of the present invention;

[0083] Fig. 9 It is another structural schematic diagram of a resource scheduling device based on business measurement provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] In content distribution networks, global load balancing is usually achieved through intelligent DNS, which distributes data to the nearest cache server that can be requested based on the LDNS address configured by the customer. Therefore, the scheduling accuracy of CDN needs to rely on the intelligent resolution of GSLB.

[0086] The current mainstream GSLB scheduling method is to forward the scheduling through the local LDNS or through the CNAME alias method.

[0087] The current scheduling strategy simply calculates the network bandwidth and server availability. It is not suitable for the personalized scheduling of VR / AR / XR and other services under the new situation. There is no dynamic weight scheduling method based on the linkage between end-to-end service quality expectation indicators and resource load expectations.

[0088] In addition, some CDN service providers rent IDC resources in adjacent areas, that is, other areas, at low prices to provide services to users in their own area in order to save costs. This leads to the following two problems: first, the scheduling delay between different areas is large, which reduces the user experience and satisfaction; second, the traffic penetrates between backbone networks, wasting the bandwidth and resources between backbone networks. This is a wrong scheduling in business management, and is not suitable for the traffic management strategy under the new model of the transformation of computing power level from centralized to "cloud-edge-end" three-dimensional.

[0089] The resource scheduling accuracy and efficiency of existing solutions are low, and the user business service quality is low.

[0090] This application proposes a resource scheduling method based on business measurement, see Figure 1 , is a flow chart of a resource scheduling method based on business measurement provided by an embodiment of the present invention, the method comprising steps S1 to S5;

[0091] S1, collects scheduling resource data for traffic service scheduling in this area;

[0092] S2, analyzing the service quality and traffic flow direction of the scheduling resource data to determine evaluation indicators;

[0093] S3, predicting the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and scoring them respectively;

[0094] S4, determining the allocation weight of the local IDC resolution fast table according to the scoring result;

[0095] S5, selecting the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address.

[0096] During the specific implementation of this embodiment, a traffic flow analysis and decision-making system is established to perform resource scheduling management. The advantage of the operator mastering LDNS is fully utilized, and part of the scheduling authority is placed next to LDNS. First, the scheduling resource data of the traffic service scheduling in this area is collected. The traffic service scheduling data in this area is the data of the business resource scheduling in this area, including the resource scheduling within the area and the traffic resource scheduling between the area and other areas. The scheduling data is collected through the regional network port to collect business traffic. The operator's CDN global scheduling system GSLB (if any) is synchronized with the scheduling view of this area, and end-to-end service quality analysis, bandwidth utilization analysis, business analysis, cross-regional analysis and other operations are performed to determine the evaluation indicators of the current resource scheduling; when performing evaluation prediction, the end-to-end service quality and bandwidth utilization of the next cycle are predicted according to the preset model.

[0097] According to the prediction results, the service quality and link bandwidth utilization of key websites are ranked, and different indicators are scored separately, and priority scheduling rights are granted.

[0098] A weighted IDC resolution address fast table is constructed based on the scoring results, and a weighted local IDC resolution fast table is deployed in the LDNS. The data in this fast table is obtained from the traffic flow analysis and decision-making system. If the weight is 0, scheduling is performed from the synchronized GSLB, thereby completing local efficient, fast, and prioritized precise resource scheduling.

[0099] This application completes the optimal scheduling through the local LDNS controlled by the operator, solves the problem that resource business scheduling cannot be blindly scheduled, and improves resource scheduling efficiency and accuracy and business service quality by performing business measurement and traffic quality analysis.

[0100] In another embodiment provided by the present invention, when collecting the scheduling resource data of the traffic service in the local area, the following steps are specifically performed:

[0101] See also Figure 2 , is another flow chart of a resource scheduling method based on business measurement provided in an embodiment of the present invention.

[0102] When the traffic flow analysis decision model is used for resource scheduling management, it is mainly divided into four layers: collection layer, analysis layer, prediction layer, and application layer.

[0103] The collection layer collects GSLB scheduling data, DNS-related IDC IP resolution data, regional network port export DPI traffic data, and IDC export DPI traffic data;

[0104] After data collection, synchronize the GSLB scheduling data;

[0105] The acquired data is aggregated, formatted, and normalized to synthesize the acquired data into an XDR call record.

[0106] In another embodiment provided by the present invention, when the step S2 is implemented, the following steps are specifically performed:

[0107] The drill-down levels of different IDC customers, different services, and different domain names are determined according to the customer information, IP address library information, and domain name information in the scheduling resource data.

[0108] See also Figure 3 , is a schematic diagram of the resource level framework for resource data collection provided by an embodiment of the present invention. For the Internet, there are only three types of resources: customer information, IP address library and domain name library. Customer information is the correspondence between IDC customers and IDC network segments. For some customers, basic annotation information can be called SP; IP address library information is the correspondence between IP and its location, that is, which operator, which region, and which computer room an IP belongs to; domain name information refers to the correspondence between domain names and business subcategories, which are divided into first-level domain names and second-level domain names. The correspondence only provides first-level domain names. The business subcategory to which the second-level domain name belongs automatically inherits the first-level domain name. The first-level domain name is XXX.COM|XXX.CN|XXX.ORG, etc., and the second-level domain name is XXX.XXX.COM, etc.

[0109] When querying domain names, you need to list all domain names. When querying business subcategories, only pay attention to the subcategories that have provided corresponding relationships. For SPs, you can drill down to the first-level domain name and the IP (that is, which IPs the SP has purchased from IDC. You can switch between displaying IPs and network segments); for first-level domain names, you can drill down to the second-level domain name and the IP (and which IPs the first-level domain name can resolve to. You only need to display the specific IP, not the network segment.

[0110] By analyzing the specific business quality and domain name quality at different drill-down levels, the business quality indicators are determined. By analyzing the specific business and domain name quality, relevant trends are found and subsequently sent to the prediction layer for decision-making.

[0111] Determine the bandwidth utilization of IDC-related links by analyzing the service traffic flow rates of websites at different drill-down levels.

[0112] The service quality index and the bandwidth utilization rate are used as the evaluation indexes.

[0113] When conducting business analysis, the analysis layer completes cross-regional traffic flow analysis, business analysis, bandwidth utilization analysis, and business quality analysis of IDC key business customers, and determines business quality indicators and the bandwidth utilization as the evaluation indicators.

[0114] In another embodiment provided by the present invention, when determining the service quality indicator, the following steps are specifically performed:

[0115] Based on the identification of domain names and IP addresses of different levels in the scheduling resource data, the drill-down levels at different levels are determined.

[0116] See also Figure 4 , is a flowchart of the quality analysis and collection process provided by the embodiment of the present invention. It is necessary to analyze the specific business and domain name quality, specifically, determine the business category dimension of the SP drill-down through the business category identification code table, perform data association, and determine the quality index of each category;

[0117] It should be noted that the quality indicators of each major category are different. The specific indicators can be input through the EXCEL of quality indicators to complete the determination of quality indicators.

[0118] After the indicators are determined, you can drill down to the first level to focus on the quality indicators of upstream and downstream traffic, including one of the following:

[0119] The IP address database is used to associate data with the IPs drilled down by the SP, analyze high-quality and low-quality IPs, and mark their secondary domain names. The IP's HTTP response success rate, HTTP response delay, TCP establishment success, and TCP establishment delay are determined as business indicators and transmitted to the prediction layer for decision-making.

[0120] The preset domain name library and the IP address library are used to associate data with the second-level domain names under the SP, and high-quality domain names and low-quality domain names are analyzed. The SP and IP to which they belong are marked in the second-level domain name data. The HTTP response success rate, HTTP response delay, TCP establishment success, and TCP establishment delay of the second-level domain name are determined as business indicators and transmitted to the prediction layer for decision-making.

[0121] It should be noted that the quality indicators in this case can be determined by setting different indicator definitions and indicator algorithms for data from different sources and indicators of different business types. For example, for web traffic business of DPI data type, the page response success rate is set as the indicator, and the probability of a user initiating web browsing to successfully establishing a web page connection is determined as the indicator definition. The indicator algorithm then obtains the product of the TCP connection establishment success rate and the first GET request success rate as the page response success rate. Indicators of other data sources and business types are not described in detail here.

[0122] In another embodiment provided by the present invention, when determining the bandwidth utilization of the IDC-related links, the following steps are specifically performed:

[0123] See also Figure 5, is a flow chart of traffic flow analysis provided by an embodiment of the present invention. When performing traffic flow analysis to determine bandwidth utilization, pay attention to the peak flow rate of each primary and secondary domain name of key websites to each region. IDC SP serves the region, other regions, this network, other downstream peak flow rates, load conditions, etc. Specifically associated with IP.

[0124] Specifically, only the downstream is focused on under the service traffic, that is, the SP that drills down in the TOP55 dimension drills down to the first-level domain name, associates the preset IP address library and the first-level domain name with data, and generates the peak flow rate and load conditions of each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers; that is, other downstream peak flow rates and load conditions of IDC SP serving this region, other regions, this network and other service providers.

[0125] Furthermore, on the basis of focusing only on the downstream, drill down and focus on the upstream and downstream, drill down the SP in the TOP55 dimension to the first-level domain name, and further drill down to the second-level domain name, and associate the IP address library with the second-level domain name drilled down from the first-level domain name to generate the peak flow rate and load conditions of each second-level domain name under each first-level domain name of the important website to each region, as well as other downstream peak flow rates of IDCSP serving other regions and other customers; that is, other downstream peak flow rates and load conditions of IDC SP serving this region, other regions, this network and other service providers.

[0126] Determine the bandwidth utilization of IDC-related links based on traffic analysis.

[0127] In another embodiment provided by the present invention, when predicting and scoring the service quality indicator, the following steps are specifically performed:

[0128] Since the regression model is ineffective for the factors that affect it, and there is no way to obtain the data information of the relevant factors, the data is the value of the time series, which may show a certain steady state or regularity in time. The triple exponential smoothing method is the best method to use the values ​​of the past time interval to predict the future values ​​and consider the periodic factors. Therefore, the triple exponential smoothing method is used to predict the indicators of the next period or the next few periods.

[0129] Calculate the bandwidth utilization and service quality indicators collected during this period to determine the stability value, trend value and seasonality value;

[0130] The average value of the flow collected from the first to the i-th time in the cycle is the stability value, denoted as s i ,s i =αx i / p i-k +(1-α)(s i-1 +ti-1 ).

[0131] The trend value is determined by the increasing or decreasing trend of the current value calculated based on the difference between the i-th and i-1-th times, denoted as t i , t i =β(s i -s i-1 )+(1-β)t i-1 .

[0132] The seasonal value is the overall increasing or decreasing trend obtained from the data collected from the first to the i-th time, denoted as p i , p i =γx i / s i +(1-γ)p i-k .

[0133] Parameter x i Represents the data collected for the i-th time in the original traffic data sequence. i+h It represents the predicted data value. The subscript i+h represents the predicted value h time intervals from the i-th data collection. When h is 1, it represents the predicted value of the indicator data collected next time. k is the number of data collections in one indicator data cycle. i-k is with p i The corresponding periodic value of the previous period. i-k+(hmodk) The physical meaning of is the seasonal value of the previous cycle of sequence number i+h.

[0134] α, β, and γ are independent variables between 0 and 1, which mainly determine the predicted value of future data. The closer they are to 1, the greater the influence of recent data values ​​on the predicted value. These independent variables can be obtained by inputting the continuous actual values ​​of the collected data.

[0135] The data input is: the associated IP bandwidth utilization, the associated IP service indicators, and the bandwidth utilization and service quality indicators are calculated separately to determine the stability value, trend value and seasonal value of the bandwidth utilization, as well as the stability value, trend value and seasonal value of the service quality indicators.

[0136] The triple exponential smoothing method is used to calculate the stability value, trend value and seasonal value of bandwidth utilization and service quality indicators respectively, and predict the bandwidth utilization rate of IP in the next cycle and the related service indicators of IP in the next cycle;

[0137] Predicted IP bandwidth utilization rate for the next cycle, predicted IP-related service indicators for the next cycle, x i+h =(s i +ht i ) i-k+(hmodk) .

[0138] After obtaining the prediction index for the next cycle, the bandwidth utilization rate of the calculated IP in the next cycle and the related business index of the IP in the next cycle are scored according to the index scoring range. Specifically:

[0139] The business prediction indicators are divided into five ranges and scored from high to low. For example, the page response success rate is 98% to 100% as 100, 95% to 97.99% as 80, 94.99% to 92% as 60, 91.99% to 90% as 40, 89.99% to 87% as 20, and below 86.99% indicates serious quality deterioration and is scored as 0.

[0140] Related IP predicted service quality score: = SUM (service response success rate score + page response delay score + ...) / number of service quality indicators / 100.

[0141] The predicted IP bandwidth utilization is divided into five ranges and scored from high to low. For example, a utilization rate of 0-10% is 100, 11%-20% is 80, 21%-30% is 60, 31%-40% is 40, 41%-50% is 20, and above 51% indicates that the utilization rate is high and is scored as 0.

[0142] Predicted bandwidth utilization score: = utilization score / 100.

[0143] In another embodiment provided by the present invention, the bandwidth utilization rate and the service quality index score of the same IP are weighted and summed to determine the weight of each IP of the domain name; that is, the weight of each IP of the same domain name: =ROUND((50*relevant IP predicted bandwidth utilization score+50*relevant IP predicted service quality score) / 10)*10 / sum of weights of the same domain name, 1)*100;

[0144] The IP weights of all IPs under the same domain name are summed up, and the sum of the weights of the same domain name is: = SUM (same domain name weight 1 + same domain name weight 2 + ...), so as to obtain the weight of each domain name and update the allocated weights of different domain names in the local IDC resolution table.

[0145] As a preferred embodiment, domain name AA has two corresponding resolution A records: IP1, IP2. IP1 predicted bandwidth utilization score is 0.80, predicted service quality score is 1; IP2 predicted bandwidth utilization score is 0.60, predicted service quality score is 0.80; then: IP1 domain name weight = ROUND (90 / 160, 1) * 100 = 60; IP2 domain name weight = ROUND (70 / 160, 1) * 100 = 40.

[0146] Record each domain name, refer to Table 1, and determine the allocation weight of the local IDC resolution fast table.

[0147] Table 1 Domain Name Record Table

[0148]

[0149] The prediction layer mainly performs triple exponential smoothing prediction on the normalized data to obtain the results of service quality, bandwidth utilization, etc. for the next cycle or the next few cycles, thereby dynamically obtaining the weight structure of the local IDC resolution fast table.

[0150] In another embodiment of the present invention, when step S5 specifically determines the resource scheduling address, see Figure 6 , is a business process diagram of a resource scheduling method based on business measurement provided in an embodiment of the present invention.

[0151] When performing a resource scheduling query, perform the following steps:

[0152] Step 1: The DPI device collects data from router nodes and IDC nodes, collects data from regional network exits and IDC exits, synthesizes XDR call records, and sends them to the traffic flow analysis and decision-making system.

[0153] Step 2: The DNS resolution log of the IDC-related IP in this area is sent to the traffic flow analysis and decision-making system.

[0154] Step 3: If it is a carrier CDN, synchronize the GSLB scheduling data of this region to the traffic flow analysis and decision-making system.

[0155] Step 4: The traffic flow analysis and decision system measures the business quality and bandwidth utilization of key websites based on the above data, predicts the situation in the next cycle and scores it. If there is no change in the next cycle, the current weight scheduling allocation result is maintained. If there is a change in the next cycle, the current scheduling weight allocation result is updated and synchronized to the local IDC resolution fast table in this area. The local IDC resolution fast table in this area is a separate process cache area of ​​​​LDNS.

[0156] Step 5: The user visits a key website and the domain name reaches the local LDNS.

[0157] Step 6: The local LDNS first searches the local IDC resolution table in the region.

[0158] Step 7: If the local IDC in the region finds a result in the fast table and the weight is not 0, it is fed back to the LDNS.

[0159] Step 8: The local LDNS feeds back the resolution result to the user.

[0160] Step 9: If the LDNS is not found in the local IDC resolution table in this area, or the weight is 0 or there is no record, then all cache areas are searched. If found, the resolution result is fed back to the user; if not found, a recursive search is performed on the authorized DNS.

[0161] Step 10: Authorize the DNS to feedback the address of the CDN domain name alias.

[0162] Step 11: The local LDNS dispatches data to the GSLB for resolution and search.

[0163] Step 12: GSLB feeds back the nearest resource address to LDNS, and then performs step 7 to feed back to the user.

[0164] In specific implementation, scheduling queries are performed preferentially in the local IDC resolution fast table.

[0165] Define websites that are not within the distribution range of the operator's CDN but whose resources are in the operator's local IDC. To prevent incorrect scheduling, establish an error correction mechanism. For example, analyze the inbound and outbound network traffic of each service during the late busy hours. The collection points are the DPI traffic of the Internet regional network port outlet and the inbound and outbound traffic of the IDC outlet. This is used to analyze the situation where the TOP55 dimension website is incorrectly scheduled across regions when there are resources in the IDC. By establishing an error correction mechanism, the service quality of IDC-related IPs is counted. If the resource occupancy rate is low and the service quality is high, the score is calculated based on the prediction results of the triple exponential smoothing method, and the weight is assigned. A local IDC resolution fast table is established in this area. The fast table records the A address and weight corresponding to the website domain name, directly completing the scheduling of large traffic resources, shortening the delay, and preventing cross-regional scheduling. If the service quality drops to a certain proportion, determine whether to go to GSLB scheduling based on the weight sorting results.

[0166] Second, if resources are scheduled in the operator's CDN, an automatic matching weight scheduling method is constructed based on the IDC resource situation, load situation, and service quality situation. The forward-looking indicator prediction is completed using the triple exponential smoothing method, the weight ratio is completed dynamically, and it is updated periodically, injected into the parsed fast table and synchronized with the GSLB data. If it is consistent with the GSLB data, it is scheduled according to the parsed fast table data. If it is inconsistent and the weight has not dropped to 0, it is scheduled according to the fast table and synchronized to the GSLB. If the weight drops to 0, it is scheduled according to the GSLB result.

[0167] The traffic flow analysis and decision-making system dynamically adjusts the distribution strategy of user requests based on the calculated weights. Figure 7 , is a schematic diagram for comparing resource scheduling processes provided in an embodiment of the present invention.

[0168] This solution dynamically adjusts the distribution strategy of user requests based on the calculated weights. For example, for nodes with heavy loads and reduced service quality, their request distribution weights are reduced, while for nodes with light loads and better service quality, their request distribution weights are increased.

[0169] By using the LDNS controlled by the operator as the scheduling function, the local IDC resolution fast table of this network is deployed on the LDNS side. If it is the operator of this network, it is synchronized with GSLB, which solves the problem of incorrect cross-regional scheduling when some local resources exist and are available in CDN operation, reduces Internet latency, and improves customer perception.

[0170] This application is aimed at the current CDN content scheduling. Some existing DNS do not support EDNS, and the source address of the user visiting the website will not be transmitted to GSLB for resolution; some CDN manufacturers have implemented EDNS, which can obtain the user's source IP, and resources can be scheduled nearby, but after the TTL, each access needs to access the authorized server and then perform the scheduling process; some CDN manufacturers that have implemented EDNS have solved the problem of erroneously scheduling resources in adjacent areas to users in this area in order to reduce IDC rental costs or misconfigure the source address and nearby resource matching table, and directly prioritize local resources. This case is based on the fact that the actual scheduling results of GSLB's global scheduling deviate from the preset weight strategy, and that users of LDNS requests may be subject to large-scale fluctuations in traffic due to frequent address switching when the cache fails each time. Weight scheduling can be implemented predictably.

[0171] The present invention does not need to modify recursive DNS and GSLB, does not need to add fields, only completes weight calculation in the traffic flow analysis and decision system, performs fast table modification on LDNS, and improves the resolution speed. The situation of frequent weight scheduling of GSLB with high load will not occur.

[0172] See also Figure 8 , is a schematic diagram of a resource scheduling device based on business measurement provided by an embodiment of the present invention, the device comprising:

[0173] The data collection module is used to collect the scheduling resource data of the traffic service scheduling in this area;

[0174] An analysis module, used to analyze the service quality and traffic flow of the scheduling resource data and determine evaluation indicators;

[0175] A prediction module, used to predict the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and score them respectively;

[0176] The weight module is used to determine the allocation weight of the local IDC resolution fast table according to the scoring results;

[0177] The scheduling module is used to select the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address.

[0178] It should be noted that the resource scheduling device based on business measurement provided in this embodiment can execute all the steps and functions of the resource scheduling method based on business measurement provided in any of the above embodiments, and the specific functions of the device are not described in detail here.

[0179] See also Fig. 9 , is another structural diagram of a resource scheduling device based on business measurement provided by an embodiment of the present invention. The resource scheduling device based on business measurement includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a resource scheduling program based on business measurement. When the processor executes the computer program, the steps in each of the above-mentioned resource scheduling method embodiments based on business measurement are implemented, such as Figure 1 Alternatively, the processor implements the functions of each module in the above-mentioned device embodiments when executing the computer program.

[0180] Exemplarily, the computer program may be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments that can perform specific functions, and the instruction segments are used to describe the execution process of the computer program in the resource scheduling device based on business measurement. For example, the computer program may be divided into several modules, and the specific functions of each module have been described in detail in a resource scheduling method based on business measurement provided in any of the above embodiments, and the specific functions of the device will not be repeated here.

[0181] The resource scheduling device based on business calculation can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The resource scheduling device based on business calculation can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the schematic diagram is only an example of a resource scheduling device based on business calculation, and does not constitute a limitation on a resource scheduling device based on business calculation. It can include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the resource scheduling device based on business calculation can also include input and output devices, network access devices, buses, etc.

[0182] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the resource scheduling device based on business measurement, and uses various interfaces and lines to connect the various parts of the entire resource scheduling device based on business measurement.

[0183] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the resource scheduling device based on business measurement by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0184] Wherein, if the module integrated in the resource scheduling device based on business measurement is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0185] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0186] It should be noted that the computer program product provided in this embodiment can execute all the steps and functions of a resource scheduling method based on business measurement provided in any of the above embodiments, and the specific functions of the device are not described in detail here.

[0187] It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A resource scheduling method based on business measurement, characterized in that: The method comprises: Collect scheduling resource data for traffic service scheduling in this area; Analyze the service quality and traffic flow of the scheduling resource data to determine evaluation indicators; Predicting the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and scoring them respectively; Determine the allocation weight of the local IDC resolution fast table based on the scoring results; The local IDC resolution fast table is selected according to the allocation weight for scheduling query to determine the resource scheduling address.

2. A resource scheduling method based on business measurement according to claim 1, characterized in that: Collect the dispatch resource data for traffic service dispatch in this area, including: Obtain GSLB scheduling data, DNS-related IDC IP resolution data, regional network port export DPI traffic data, and IDC export DPI traffic data for the region's traffic services; Synchronously backing up the GSLB scheduling data; The acquired data is converted into a format and normalized to synthesize the acquired data into an XDR call record.

3. The resource scheduling method based on business measurement according to claim 1, characterized in that: Analyze the service quality and traffic flow of the scheduling resource data to determine evaluation indicators, including: Determine the drill-down levels of different IDC customers, different services, and different domain names according to the customer information, IP address library information, and domain name information in the scheduling resource data; Determine the service quality indicators by analyzing the specific service quality and domain name quality at different drill-down levels; Determine the bandwidth utilization of IDC-related links by analyzing the service traffic flow rates of websites at different drill-down levels; The service quality index and the bandwidth utilization rate are used as the evaluation indexes.

4. A resource scheduling method based on business measurement according to claim 3, characterized in that: Determine service quality indicators, including: Data association is performed between the preset IP address database and the IPs drilled down by SPs of major business categories, and high-quality and low-quality IPs are analyzed. The secondary domain names to which they belong are marked, and the HTTP response success rate, HTTP response delay, TCP establishment success, and TCP establishment delay of the IPs are determined as business indicators. The preset domain name library, the IP address library and the secondary domain names drilled down by the SP of the major business category are associated with each other, and high-quality domain names and low-quality domain names are analyzed. The SP and IP to which they belong are marked in the secondary domain name data, and the HTTP response success rate, HTTP response delay, TCP establishment success and TCP establishment delay of the secondary domain name are determined as business indicators.

5. The resource scheduling method based on business measurement according to claim 3, characterized in that: Determine the bandwidth utilization of IDC-related links, including: The preset IP address database is associated with the first-level domain name drilled down by the SP of the business category to generate the peak flow rate and load status of each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers; The IP address database is associated with the second-level domain name drilled down from the first-level domain name to generate peak flow rates and load conditions of each second-level domain name under each first-level domain name of important websites to each region, as well as other downstream peak flow rates of IDC SP serving other regions and other customers; Determine the bandwidth utilization of IDC-related links based on traffic analysis.

6. A resource scheduling method based on business measurement according to claim 1, characterized in that: The bandwidth utilization and service quality indicators of the next cycle are predicted based on the evaluation indicators, and are scored respectively, including: Calculate the bandwidth utilization and service quality indicators collected during this period to determine the stability value, trend value and seasonality value; The triple exponential smoothing method is used to calculate the stability value, trend value and seasonal value of bandwidth utilization and service quality indicators respectively, and predict the bandwidth utilization rate of IP in the next cycle and the related service indicators of IP in the next cycle; The calculated bandwidth utilization of the IP in the next cycle and related business indicators of the IP in the next cycle are scored respectively according to the pre-divided indicator scoring intervals.

7. The resource scheduling method based on business measurement according to claim 1, characterized in that: The allocation weight of the local IDC resolution fast table is determined based on the scoring results, including: The bandwidth utilization rate and service quality index scores of the same IP are weighted and summed to determine the weight of each IP of the domain name; The IP weights of all IPs under the same domain name are summed up, and the allocation weights of different domain names in the local IDC resolution fast table are updated.

8. The resource scheduling method based on business measurement according to claim 1, characterized in that: The local IDC resolution fast table is selected according to the allocation weight to perform scheduling query and determine the resource scheduling address, including: Searching according to the local IDC parsing quick table; When the resource address is found and the corresponding allocation weight is not 0, it is fed back to the LDNS for resource scheduling; When the resource address is not found, or the allocation weight corresponding to the found resource address is 0, the GSLB scheduling data in the scheduling resource data is parsed and searched, and the latest resource address found is fed back to the LDNS for resource scheduling.

9. A resource scheduling device based on business measurement, characterized in that: The device comprises: The data collection module is used to collect the scheduling resource data of the traffic service scheduling in this area; An analysis module, used to analyze the service quality and traffic flow of the scheduling resource data and determine evaluation indicators; A prediction module, used to predict the bandwidth utilization and service quality index of the next cycle according to the evaluation index, and score them respectively; The weight module is used to determine the allocation weight of the local IDC resolution fast table according to the scoring results; The scheduling module is used to select the local IDC resolution fast table for scheduling query according to the allocation weight to determine the resource scheduling address.

10. A resource scheduling device based on business measurement, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a resource scheduling method based on business measurement as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute a resource scheduling method based on business measurement as described in any one of claims 1 to 8.

12. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method described in any one of claims 1 to 8 are implemented.

Images