Traffic diversion scheduling method, device, system and computer-readable storage medium
By dynamically adjusting the shunt scheduling strategy using the traffic prediction model in DPI equipment, the problem that the scheduling strategy in the existing technology cannot be adjusted according to traffic changes is solved, and server utilization and system efficiency are improved.
Patent Information
- Application Number
- CN201910787404.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-08-26
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2039-08-26
AI Technical Summary
The scheduling policies in existing DPI devices cannot be dynamically adjusted according to changes in traffic, resulting in low server utilization and low system efficiency.
By obtaining the historical traffic information of the link, establishing a traffic prediction model, determining the predicted traffic information for different time periods in the future, and adjusting the shunt scheduling strategy based on the prediction information to adapt to the traffic needs of different time periods.
This improves the utilization rate of the analysis server, thereby improving the efficiency of the system, and ensuring the satisfaction of traffic requirements in different time periods.
Smart Images

Figure CN112437015B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of communication technology, and in particular to a traffic diversion scheduling method, a traffic diversion scheduling device, a traffic diversion scheduling system, and a computer-readable storage medium. Background Art
[0002] DPI (Deep Packet Inspection) devices can collect and identify traffic on the link and meet the traffic or analysis statistics required by various upper-layer applications. The diverter in the DPI device can realize functions such as large traffic splitting, same-source and same-destination, and load balancing. The diverter can evenly distribute the traffic to the analysis server for processing according to the configured diversion strategy.
[0003] In the related art, when a DPI device is deployed, a fixed diversion strategy is configured according to the peak traffic of the link to schedule the diverter, and a fixed number of analysis servers are configured to perform deep packet inspection on the traffic. Summary of the invention
[0004] The inventors of the present disclosure have discovered that the above-mentioned related technologies have the following problems: the scheduling strategy cannot be dynamically adjusted according to changes in traffic, resulting in low server utilization and thus low system efficiency.
[0005] In view of this, the present disclosure proposes a diversion scheduling technical solution that can improve system efficiency.
[0006] According to some embodiments of the present disclosure, a diversion scheduling method is provided, including: obtaining historical traffic information in a link; establishing a traffic prediction model based on the historical traffic information; using the traffic prediction model to determine predicted traffic information in different time periods in the future; and determining a diversion scheduling strategy in different time periods in the future based on the predicted traffic information.
[0007] In some embodiments, establishing a traffic prediction model based on the historical traffic information includes: fitting a polynomial regression function relationship between each moment and the corresponding traffic size as the traffic prediction model based on the historical traffic information.
[0008] In some embodiments, the diversion scheduling method further includes: verifying whether the diversion scheduling strategy can meet the actual traffic demand based on actual traffic information in different time periods in the future; if the actual traffic demand cannot be met, adjusting the diversion scheduling strategy according to the actual traffic information.
[0009] In some embodiments, the acquiring of historical traffic information of a link includes acquiring historical outflow peak traffic and historical inflow peak traffic at each moment in the link as historical traffic information.
[0010] In some embodiments, the obtaining of historical traffic information of a link includes: obtaining the historical traffic information via a simple network management protocol.
[0011] In some embodiments, determining the diversion scheduling strategy in different future time periods based on the predicted traffic information includes: determining the number of analysis servers based on the predicted traffic information and the traffic processing capacity of each analysis server.
[0012] In some embodiments, the predicted traffic information includes predicted traffic sizes at each future moment, and the different future time periods are divided according to the predicted traffic sizes at each future moment.
[0013] According to some other embodiments of the present disclosure, a diversion scheduling device is provided, including: an acquisition unit, used to acquire historical traffic information in a link; an establishment unit, used to establish a traffic prediction model based on the historical traffic information; a prediction unit, used to determine the predicted traffic information in different time periods in the future using the traffic prediction model; and a determination unit, used to determine the diversion scheduling strategy in different time periods in the future based on the predicted traffic information.
[0014] In some embodiments, the establishing unit fits a polynomial regression function relationship between each moment and the corresponding flow size according to the historical flow information as the flow prediction model.
[0015] In some embodiments, the diversion scheduling device further includes: a verification unit, used to verify whether the diversion scheduling strategy can meet the actual traffic demand based on the actual traffic information in different time periods in the future; and an adjustment unit, used to adjust the diversion scheduling strategy according to the actual traffic information when the actual traffic demand cannot be met.
[0016] In some embodiments, the acquisition unit acquires the historical outflow peak flow and the historical inflow peak flow at each moment in the link as the historical flow information.
[0017] In some embodiments, the acquisition unit acquires the historical traffic information via a simple network management protocol.
[0018] In some embodiments, the determination unit determines the number of analysis servers based on the predicted traffic information and the traffic processing capabilities of each analysis server.
[0019] In some embodiments, the predicted traffic information includes predicted traffic sizes at each future moment, and the different future time periods are divided according to the predicted traffic sizes at each future moment.
[0020] According to some further embodiments of the present disclosure, a traffic diversion scheduling device is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the traffic diversion scheduling method in any one of the above embodiments based on instructions stored in the memory device.
[0021] According to some further embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the offload scheduling method in any of the above embodiments is implemented.
[0022] According to some further embodiments of the present disclosure, a diversion scheduling system is provided, comprising the diversion scheduling device in any one of the above embodiments.
[0023] In some embodiments, the diversion scheduling system further includes: a diverter for sending the traffic to be analyzed to the analysis server according to the diversion scheduling strategy determined by the diversion scheduling device.
[0024] In the above embodiment, a traffic prediction model is established based on historical traffic information, and the analysis server is scheduled in different time periods based on the traffic prediction results. In this way, the splitter can be dynamically scheduled according to the traffic changes in different time periods to adapt to the traffic requirements in different time periods, thereby improving the utilization rate of the analysis server and further improving the efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[0026] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:
[0027] Figure 1 A flowchart showing some embodiments of the diversion scheduling method disclosed in the present invention;
[0028] Figure 2 A schematic diagram showing some embodiments of the diversion scheduling method disclosed in the present invention;
[0029] Figure 3 Flow charts showing other embodiments of the diversion scheduling method disclosed in the present invention;
[0030] Figure 4 A block diagram showing some embodiments of the diversion scheduling device disclosed in the present invention;
[0031] Figure 5 A block diagram showing some other embodiments of the diversion scheduling device disclosed in the present invention;
[0032] Figure 6 A block diagram showing some further embodiments of the diversion scheduling device disclosed in the present invention;
[0033] Figure 7 A block diagram showing some embodiments of the diversion scheduling system of the present disclosure. DETAILED DESCRIPTION
[0034] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.
[0035] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.
[0036] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.
[0037] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.
[0038] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.
[0039] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.
[0040] Figure 1 A flow chart showing some embodiments of the diversion scheduling method disclosed in the present invention.
[0041] like Figure 1 As shown, the method includes: step 110, obtaining historical traffic information; step 120, establishing a traffic prediction model; step 130, determining the predicted traffic information; and step 140, determining a diversion scheduling strategy.
[0042] In step 110, historical traffic information in the link is obtained. For example, the historical traffic information can be obtained through SNMP (Simple Network Management Protocol).
[0043] In some embodiments, the historical outflow peak flow and the historical inflow peak flow at each moment in the link are obtained as the historical flow information. For example, the uplink and downlink peak flows (such as the sum of the historical outflow peak flow and the historical inflow peak flow) of the link per hour in the past period of time (such as 7 days or 30 days) can be obtained through SNMP as the historical flow information.
[0044] In this way, the traffic size of the link at each historical moment can be obtained as the basis for dynamic scheduling of each analysis server in the future, so as to avoid calling too many or too few analysis servers (for example, under normal circumstances, the peak traffic of the link is about 3 times the trough traffic, and setting a fixed scheduling strategy according to the peak traffic will cause waste of analysis server resources), thereby improving the utilization rate of each analysis server.
[0045] In step 120, a traffic prediction model is established based on historical traffic information. Since the curve change of traffic in the link presents a regular state (such as 4 o'clock every day is the traffic trough, 20 o'clock is the traffic peak, etc.), the traffic prediction model established based on historical traffic information can serve as a reliable basis for future dynamic scheduling of splitters and analysis servers.
[0046] In some embodiments, a polynomial regression function relationship between each moment and the corresponding flow size can be fitted as a flow prediction model based on historical flow information. For example, the following polynomial regression function can be established as a flow prediction model:
[0047]
[0048] x can be any time of the day, such as x=1 means 1 o'clock of the day, etc.; y is the flow rate corresponding to time x; α m are polynomial coefficients, α m It can be a positive or negative number; M is the degree of the polynomial, M is an integer greater than or equal to 0; m is an integer greater than or equal to 0 and less than or equal to M.
[0049] In some embodiments, the sampling values of x for 24 hours a day in the past 7 days can be taken, and the corresponding flow conditions (outflow peak + inflow peak) can be taken as the sampling values of y. The curve of f(x) is fitted based on these sampling values to obtain the flow prediction model. For example, M=6 can be taken, and the polynomial coefficients α can be determined based on each sampling value. m , so the traffic prediction model is:
[0050] y=f(x)=-0.0999x 6 +6.1111x 5 -131.26x 4 +1088.6x 3 -1314.6x 2 -14643x+70211
[0051] In step 130, the traffic prediction model is used to determine predicted traffic information in different time periods in the future.
[0052] In some embodiments, the predicted traffic information at any time in the future can be obtained according to the above traffic prediction model. For example, the traffic prediction model f(x) obtained according to the above embodiment predicts the traffic size of the link at any time on the second day. For example, x=7, y=52485.2626, which means that the predicted traffic size of the link at 7 o'clock on the second day is 52485.2626Mbps.
[0053] In step 140, the diversion scheduling strategies for different time periods in the future are determined based on the predicted traffic information.
[0054] In some embodiments, the number of analysis servers can be determined based on the predicted traffic information and the traffic processing capacity of each analysis server. For example, if the processing capacity of each analysis server is 10 Gbps, and the predicted traffic size of the link at 7 o'clock the next day is 52485.2626 Mbps, the quotient of the traffic size and the processing capacity can be rounded up to determine that 6 analysis servers need to be scheduled at 7 o'clock the next day.
[0055] In some embodiments, the determined number of analysis servers is sent to the diverter so that the diverter sends the traffic to be analyzed to the corresponding number of analysis servers. In some embodiments, other analysis servers other than the analysis servers involved in the diversion scheduling strategy can be controlled to sleep, thereby achieving energy saving and emission reduction.
[0056] In some embodiments, the predicted traffic information includes the predicted traffic size at each moment in the future, and different time periods in the future are divided according to the predicted traffic size at each moment. For example, adjacent moments whose predicted traffic size difference is less than a threshold (i.e., the number of analysis servers required is similar) can be divided into one time period (i.e., planning multiple time windows).
[0057] For example, according to the predicted traffic information, it is determined that the traffic size from 1:00 to 7:00 is similar; the traffic size from 8:00 to 17:00 is similar; the traffic size from 18:00 to 24:00 is similar. In this case, the above-mentioned moments can be divided into a time period, and the maximum number of analysis servers required at each moment in the time period is taken as the number of analysis servers required for the time period.
[0058] For example, the first time period includes 1:00 to 7:00, and 6 analysis servers are required; the second time period includes 8:00 to 17:00, and 9 analysis servers are required; the third time period includes 18:00 to 24:00, and 13 analysis servers are required.
[0059] In this way, a day can be divided into multiple time periods for differentiated scheduling. On the one hand, compared with 24-hour differentiated scheduling, the scheduling efficiency can be improved; on the other hand, the appropriate number of analysis servers can be scheduled according to the traffic demand in different time periods, thereby improving the utilization rate of the analysis servers.
[0060] Figure 2 Schematic diagrams showing some embodiments of the diversion scheduling method disclosed in the present invention.
[0061] like Figure 2 As shown, a DPI (Deep Packet Inspection) device may include an optical splitter (or bypass device), a flow splitter, and an analysis server (such as n units). The DPI device can collect and identify traffic on the link and meet the traffic or analysis statistics required by various upper-layer applications.
[0062] In some embodiments, the diversion scheduling device (such as a dynamic diversion scheduling server) in any of the above embodiments can be deployed to establish a prediction model based on historical traffic conditions to calculate the traffic conditions of the next day. Based on the traffic conditions of the next day, the diversion scheduling device can determine the diversion scheduling strategy of each analysis server and send it to the diverter for diversion processing. The diversion scheduling device can also continuously verify and optimize the diversion scheduling strategy according to the actual traffic conditions of the next day.
[0063] For example, the diversion scheduling device can determine the diversion scheduling strategy through the following steps:
[0064] 1) Obtain link traffic (such as the peak upstream and downstream traffic per hour in the past 7 or 30 days) through SNMP and establish a traffic prediction model;
[0065] 2) Calculate the traffic situation for the next day based on the traffic prediction model;
[0066] 3) Determine the diversion scheduling strategy. For example, you can divide the busy period and the idle period, calculate the number of analysis servers required for each period, and send the diversion scheduling strategy to the diverter regularly;
[0067] 4) Verify the predicted flow rate based on the current actual flow rate, and continuously optimize the diversion scheduling strategy based on the verification results.
[0068] In the above embodiment, by deploying a dynamic diversion scheduling server, the diversion scheduling strategy is dynamically adjusted according to the link traffic situation, so that the ratio of the analysis server is optimized, thereby improving efficiency.
[0069] In some embodiments, Figure 3 The implementation example in optimizes the diversion scheduling strategy.
[0070] Figure 3 Flowcharts showing other embodiments of the diversion scheduling method disclosed in the present invention.
[0071] like Figure 3 As shown, the method further includes: step 310, verifying whether the actual traffic demand can be met; and step 320, adjusting the diversion scheduling strategy.
[0072] In step 310, based on the actual traffic information in different time periods in the future, it is checked whether the diversion scheduling strategy can meet the actual traffic demand. If it can be met, the original diversion scheduling strategy is used for scheduling; if it cannot be met, step 320 is executed.
[0073] In step 320, when the actual traffic demand cannot be met, the diversion scheduling strategy is adjusted according to the actual traffic information.
[0074] In some embodiments, the processing capacity of each analysis server is 10 Gbps, and the predicted traffic size of the link at 7 o'clock the next day is 52485.2626 Mbps. The diversion scheduling strategy can be determined as: scheduling 6 analysis servers; however, the actual traffic size of the link at 7 o'clock the next day is 62000 Mbps, and 6 analysis servers cannot meet the actual traffic demand, so the diversion scheduling strategy needs to be adjusted to: scheduling 7 analysis servers.
[0075] In the above embodiment, a solution for DPI dynamic diversion is proposed, by setting up a dynamic diversion server, obtaining the historical traffic of the link, establishing a dynamic diversion scheduling strategy, and sending it to the diverter in different time periods. In this way, the utilization rate of the DPI analysis server is optimized, which is helpful for energy saving and emission reduction.
[0076] In addition, the above-mentioned embodiments realize the dynamic scheduling of analysis server resources and promote the development and application of DPI systems based on SDN (Software Defined Network).
[0077] In the above embodiment, a traffic prediction model is established based on historical traffic information, and the analysis server is scheduled in different time periods based on the traffic prediction results. In this way, the diversion server can be dynamically scheduled according to the traffic changes in different time periods to adapt to the traffic demand in different time periods, thereby improving the utilization rate of the server and further improving the efficiency of the system.
[0078] Figure 4 A block diagram showing some embodiments of the flow distribution scheduling device disclosed in the present invention.
[0079] like Figure 4 As shown, the offload scheduling device 4 includes an acquisition unit 41 , an establishment unit 42 , a prediction unit 43 and a determination unit 44 .
[0080] The acquisition unit 41 acquires historical traffic information in the link. For example, the acquisition unit 41 acquires the historical outflow peak traffic and the historical inflow peak traffic at each moment in the link as the historical traffic information. The acquisition unit 41 can acquire the historical traffic information through a simple network management protocol.
[0081] In some embodiments, the acquisition unit 41 acquires the historical outflow peak flow and the historical inflow peak flow at each moment in the link as the historical flow information. For example, the acquisition unit 41 acquires the historical flow information through a simple network management protocol.
[0082] The establishing unit 42 establishes a flow prediction model according to the historical flow information. For example, the establishing unit 42 fits a polynomial regression function relationship between each moment and the corresponding flow size according to the historical flow information as the flow prediction model.
[0083] In some embodiments, the establishing unit 42 fits the polynomial regression function relationship between each moment and the corresponding flow size according to the historical flow information as the flow prediction model.
[0084] The prediction unit 43 uses the traffic prediction model to determine the predicted traffic information in different time periods in the future. For example, the predicted traffic information includes the predicted traffic size at each moment in the future, and different time periods in the future are divided according to the predicted traffic size at each moment in the future.
[0085] The determination unit 44 determines the traffic dispatching strategies for the traffic diverters and analysis servers in different time periods in the future according to the predicted traffic information. For example, the determination unit 44 determines the number of analysis servers according to the predicted traffic information and the traffic processing capabilities of the analysis servers.
[0086] In some embodiments, the determination unit 44 determines the number of analysis servers according to the predicted traffic information and the traffic processing capacity of each analysis server. For example, the predicted traffic information includes the predicted traffic size at each future moment, and different time periods in the future are divided according to the predicted traffic size at each future moment.
[0087] In some embodiments, the diversion scheduling device 4 further includes a verification unit 45 and an adjustment unit 46. The verification unit 45 verifies whether the diversion scheduling strategy can meet the actual flow demand based on the actual flow information in different time periods in the future; the adjustment unit 46 adjusts the diversion scheduling strategy based on the actual flow information when the actual flow demand cannot be met.
[0088] In some embodiments, the diversion scheduling device 4 also includes: a verification unit 45 is used to verify whether the diversion scheduling strategy can meet the actual traffic demand based on the actual traffic information in different time periods in the future; the adjustment unit 46 is used to adjust the diversion scheduling strategy according to the actual traffic information when the actual traffic demand cannot be met.
[0089] In some embodiments, the diversion scheduling device 4 can control other analysis servers other than the analysis servers involved in the diversion scheduling strategy to sleep, thereby achieving energy saving and emission reduction.
[0090] In the above embodiment, a traffic prediction model is established based on historical traffic information, and the analysis server is scheduled in different time periods based on the traffic prediction results. In this way, the diversion server can be dynamically scheduled according to the traffic changes in different time periods to adapt to the traffic demand in different time periods, thereby improving the utilization rate of the server and further improving the efficiency of the system.
[0091] Figure 5 A block diagram showing some other embodiments of the flow distribution scheduling device disclosed in the present invention.
[0092] like Figure 5 As shown, the diversion scheduling device 5 of this embodiment includes: a memory 51 and a processor 52 coupled to the memory 51 , and the processor 52 is configured to execute the diversion scheduling method in any one of the embodiments of the present disclosure based on the instructions stored in the memory 51 .
[0093] The memory 51 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, a database, and other programs.
[0094] Figure 6 A block diagram showing some further embodiments of the flow distribution scheduling device disclosed in the present invention.
[0095] like Figure 6 As shown, the traffic diversion scheduling device 6 of this embodiment includes: a memory 610 and a processor 620 coupled to the memory 610 , and the processor 620 is configured to execute the traffic diversion scheduling method in any one of the aforementioned embodiments based on the instructions stored in the memory 610 .
[0096] The memory 610 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.
[0097] The flow distribution and scheduling device 6 may also include an input / output interface 630, a network interface 640, a storage interface 650, etc. These interfaces 630, 640, 650, the memory 610, and the processor 620 may be connected, for example, via a bus 660. The input / output interface 630 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 640 provides a connection interface for various networked devices. The storage interface 650 provides a connection interface for external storage devices such as SD cards and USB flash drives.
[0098] Figure 7 A block diagram showing some embodiments of the diversion scheduling system of the present disclosure.
[0099] like Figure 7 As shown, the flow scheduling system 7 includes the flow scheduling device 71 in any of the above embodiments. In some embodiments, the flow scheduling system 7 also includes a flow splitter 72 for sending the flow to be analyzed to the analysis server according to the flow scheduling strategy determined by the flow scheduling device 71.
[0100] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0101] So far, the shunt scheduling method, shunt scheduling device, shunt scheduling system and computer-readable storage medium according to the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed herein.
[0102] The method and system of the present disclosure may be implemented in many ways. For example, the method and system of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
[0103] Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.
Claims
1. A traffic diversion scheduling method, comprising: Obtain historical traffic information in the link; Establishing a traffic prediction model based on the historical traffic information; Using the traffic prediction model, determining predicted traffic information at different time periods in the future; Determine the diversion scheduling strategy in different time periods in the future according to the predicted traffic information; Wherein, establishing a traffic prediction model according to the historical traffic information includes: According to the historical traffic information, a polynomial regression function relationship between each moment and the corresponding traffic size is fitted as the traffic prediction model, wherein the polynomial regression function relationship is an M-order polynomial regression function relationship, where M is a positive integer greater than 1. in, The predicted traffic information includes predicted traffic sizes at each future moment, and the different future time periods are divided according to the predicted traffic sizes at each future moment, and adjacent future moments whose predicted traffic sizes differ by less than a threshold are divided into one time period.
2. The offload scheduling method according to claim 1, further comprising: According to the actual flow information in different time periods in the future, verify whether the diversion scheduling strategy can meet the actual flow demand; When the actual traffic demand cannot be met, the diversion scheduling strategy is adjusted according to the actual traffic information.
3. The offload scheduling method according to claim 1, wherein: The acquisition of historical traffic information of a link includes: The historical outflow peak flow rate and the historical inflow peak flow rate at each moment in the link are obtained as the historical flow information.
4. The traffic diversion scheduling method according to any one of claims 1 to 3, wherein: The acquisition of historical traffic information of a link includes: The historical traffic information is obtained through a simple network management protocol.
5. The traffic diversion scheduling method according to any one of claims 1 to 3, wherein: Determining the diversion scheduling strategies in different time periods in the future according to the predicted traffic information includes: The number of analysis servers is determined based on the predicted traffic information and the traffic processing capability of each analysis server.
6. A flow distribution and scheduling device, comprising: An acquisition unit, used for acquiring historical traffic information in the link; An establishing unit, used to establish a flow prediction model according to the historical flow information; A prediction unit, used to determine predicted traffic information in different time periods in the future using the traffic prediction model; A determination unit, configured to determine a diversion scheduling strategy in different time periods in the future according to the predicted traffic information; in, The establishment unit fits the polynomial regression function relationship between each moment and the corresponding flow size as the flow prediction model according to the historical flow information, wherein the polynomial regression function relationship is an M-order polynomial regression function relationship, where M is a positive integer greater than 1. The predicted traffic information includes predicted traffic sizes at each future moment, and the different future time periods are divided according to the predicted traffic sizes at each future moment, and adjacent future moments whose predicted traffic sizes differ by less than a threshold are divided into one time period.
7. The flow distribution and scheduling device according to claim 6, further comprising: A verification unit, used to verify whether the diversion scheduling strategy can meet the actual flow demand according to the actual flow information in different time periods in the future; An adjustment unit is used to adjust the diversion scheduling strategy according to the actual traffic information when the actual traffic demand cannot be met.
8. The flow distribution and scheduling device according to claim 6, wherein: The acquisition unit acquires the historical outflow peak flow and the historical inflow peak flow at each moment in the link as the historical flow information.
9. The flow distribution and scheduling device according to any one of claims 6 to 8, wherein: The acquisition unit acquires the historical traffic information through a simple network management protocol.
10. The flow distribution and scheduling device according to any one of claims 6 to 8, wherein: The determination unit determines the number of analysis servers according to the predicted traffic information and the traffic processing capability of each analysis server.
11. A flow distribution and scheduling device, comprising: Memory; and A processor coupled to the memory, wherein the processor is configured to execute the traffic distribution scheduling method according to any one of claims 1 to 5 based on instructions stored in the memory.
12. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the method for offloading and scheduling according to any one of claims 1 to 5 is implemented.
13. A diversion scheduling system, comprising: The diversion scheduling device as described in any one of claims 6 to 11.
14. The flow distribution and scheduling system according to claim 13, further comprising: The diverter is used to send the traffic to be analyzed to the analysis server according to the diversion scheduling strategy determined by the diversion scheduling device.
Citation Information
Patent Citations
Method and system for controlling service chain virtual machine
CN108616377A
Content distribution network traffic scheduling method and device and computer readable storage medium
CN109842563A
Network traffic scheduling method and device of data center
JP2017126975A
System and method of forecasting usage of network links
US20090245107A1
Network service scheduling method and apparatus, storage medium, and program product
US20180302296A1
