CN114095441A - Method for realizing ECMP flow load balance and electronic equipment - Google Patents
Method for realizing ECMP flow load balance and electronic equipment Download PDFInfo
- Publication number
- CN114095441A CN114095441A CN202111348567.4A CN202111348567A CN114095441A CN 114095441 A CN114095441 A CN 114095441A CN 202111348567 A CN202111348567 A CN 202111348567A CN 114095441 A CN114095441 A CN 114095441A
- Authority
- CN
- China
- Prior art keywords
- flow
- link
- data
- elephant
- bandwidth occupancy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2483—Traffic characterised by specific attributes, e.g. priority or QoS involving identification of individual flows
Abstract
The present specification provides a method and an electronic device for implementing ECMP traffic load balancing, where the method is applied to a forwarding device, and includes: monitoring the bandwidth occupancy rate of each link to which a plurality of data flows with the same target address are allocated, wherein the forwarding equipment allocates the links to the plurality of data flows with the same target address based on a Hash algorithm; under the condition that the bandwidth occupancy rates of the links are not balanced, identifying the types of data streams in the links of which the bandwidth occupancy rates exceed a threshold value, wherein the types of the data streams comprise elephant streams; and switching the identified elephant flow to a light-load link according to an ACL rule, wherein the ACL rule is configured according to the five-tuple information of the identified elephant flow, and the light-load link is a link of which the bandwidth occupancy rate does not exceed the threshold value. The method and the device can find the problem of unbalanced link flow in time, and realize the flow load balance of each link by switching the elephant flow on the overload link to other light-load links.
Description
Technical Field
The present disclosure relates to the field of communications technologies, and in particular, to a method and an electronic device for implementing load balancing of ECMP traffic.
Background
The multipath load balancing problem existing in a network using TCP/IP as a transport protocol is usually solved by using an ECMP (Equal Cost Multi-path) routing technology. The ECMP generally selects a HASH algorithm to determine transmission paths of different data streams, but this method only considers the balance of the number of data streams on each path when determining the transmission path, but does not consider the size of each data stream, and the problem of unbalanced traffic of each path (link) may occur.
The DLB (Dynamic load balance, Dynamic link load) technique solves the problem of traffic imbalance in the ECMP by transmitting the packets in the elephant flow in different paths. However, the method is complex to implement, needs chip support, and cannot be supported by most forwarding devices.
Disclosure of Invention
In order to overcome the problems in the related art, the present specification provides a method and an electronic device for implementing ECMP traffic load balancing.
According to a first aspect of embodiments herein, there is provided a method for implementing ECMP traffic load balancing, the method including:
monitoring the bandwidth occupancy rate of each link to which a plurality of data flows with the same target address are allocated, wherein the forwarding equipment allocates the links to the plurality of data flows with the same target address based on a Hash algorithm;
under the condition that the bandwidth occupancy rates of the links are not balanced, identifying the types of data streams in the links of which the bandwidth occupancy rates exceed a threshold value, wherein the types of the data streams comprise elephant streams;
and switching the identified elephant flow to a light-load link according to an ACL rule, wherein the ACL rule is configured according to the five-tuple information of the identified elephant flow, and the light-load link is a link of which the bandwidth occupancy rate does not exceed the threshold value.
According to a second aspect of embodiments herein, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method of the first aspect described above.
According to a third aspect of embodiments herein, there is provided an electronic apparatus including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the method of the first aspect.
The technical scheme provided by the embodiment of the specification can have the following beneficial effects:
on the basis that ECMP adopts HASH algorithm to determine the forwarding path of the data stream, the problem of unbalanced flow of each link can be found in time by monitoring the bandwidth occupancy rate of each link, the type of the data stream in the link with the bandwidth occupancy rate exceeding the threshold value is immediately identified, if the elephant stream exists, the identified elephant stream is switched to other light-load links with the bandwidth occupancy rates not exceeding the threshold value, and therefore the flow load balance of each link is achieved. In addition, the method does not need special chip support, is simple to realize and can be supported by most forwarding equipment.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.
Fig. 1 is a flowchart illustrating a method for implementing ECMP traffic load balancing according to an exemplary embodiment.
Fig. 2 is a diagram illustrating an application scenario for implementing an ECMP traffic load balancing method according to an exemplary embodiment of the present disclosure.
Fig. 3 is a flowchart illustrating a method for implementing ECMP traffic load balancing according to another exemplary embodiment.
FIG. 4 is a block diagram of an electronic device shown in accordance with an exemplary embodiment of the present description.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.
The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the description. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information, without departing from the scope of the present specification. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
ECMP routing is often applied to a network using TCP/IP as a transport protocol to solve the problem of multi-path load balancing, so as to improve network redundancy and reliability and network resource utilization, and a HASH algorithm is generally used to determine transmission paths of different data streams. However, this method only considers the balance of the number of data streams on each path, and does not consider the size of each data stream when determining the transmission path. Therefore, when a large-volume elephant flow and a small-volume mouse flow coexist, a problem of link traffic imbalance may occur. The elephant flow and the mouse flow can be distinguished through the size and the transmission rate of the data flow, and the elephant flow can be a data flow with a large transmission rate and a long transmission time, such as a data flow generated by services of data migration, virtual machine migration and the like; the mouse stream may be a data stream with a small transmission rate and a short transmission time, such as a data stream generated by a service of sending a mail, browsing a web page, and the like. Four different data streams are transmitted to the user 2 through the forwarding device and can be forwarded through two equivalent paths with the same bandwidth, wherein the data streams f1 and f3 are rat flows, and the data streams f2 and f4 are elephant flows. After HASH calculation, data streams f1 and f3 are transmitted via link 1, and account for 10% of the bandwidth of link 1; data flows f2 and f4 are transmitted over link 2, accounting for 60% of the link 2 bandwidth, thereby presenting a problem of link traffic imbalance.
The DLB dynamic link load technology distributes the data volume in the same data stream to a plurality of forwarding paths by transmitting the messages in the elephant flow in different paths, so as to solve the problem of link flow imbalance in the ECMP and balance the flow load of each path. In this process, the forwarding device needs to monitor the forwarding delays of multiple forwarding paths and the time intervals between messages in real time to prevent the data receiving end from generating message disorder. However, the method is complex to implement, needs chip support, and cannot be supported by most forwarding devices. Therefore, the application provides a method which is simple to realize, can be supported by most equipment and can balance the traffic load of each path on the basis of the ECMP technology.
The application provides a method for realizing ECMP flow load balancing in a TCP/IP network, which is applied to data forwarding equipment, such as a router or a switch. The method is applicable to the case that two or more equivalent paths exist, and the embodiment of the present specification describes the method of the present application in the case of two equivalent paths, including the following steps, as shown in fig. 1:
s110, monitoring the bandwidth occupancy rate of each link to which a plurality of data streams with the same target address are allocated, wherein the forwarding device allocates the links to the plurality of data streams with the same target address based on a Hash algorithm;
the "links" are at least two paths among the equal cost paths specified based on the same destination address, and the number of hops passed when the data flow is transferred through each of the forwarding paths in the equal cost paths is the same. When forwarding equipment such as a router or a switch forwards the data stream, the shortest forwarding path based on the hop count is obtained according to the target address, and the data stream to be forwarded is transmitted through the shortest forwarding path. When the hop counts of two or more forwarding paths are the same and the hop count is the minimum, the two or more forwarding paths are equivalent paths. When there are two or more data flows to be forwarded, even if there are two or more equal-cost paths, the forwarding device still selects one path to forward all data flows at a time, thereby causing waste of link resources. The forwarding device configured with the ECMP simultaneously allocates different data streams to be forwarded to different paths for forwarding by using all the existing equivalent paths, so that loads on different paths are balanced as much as possible. The ECMP generally determines a forwarding path of each data stream to be forwarded by using a HASH algorithm, and may also determine the forwarding path in a polling manner or based on a weight of each forwarding path. And monitoring the bandwidth occupancy rate of each link in real time, and finding out the problem of unbalanced link flow load in real time.
The data forwarding devices such as routers and switches have the function of counting traffic information at each port, different links respectively correspond to different output ports on the forwarding devices, and the respective bandwidth occupancy rates of each link are obtained by counting the data volume passing through the output port of the forwarding device corresponding to each link in unit time (such as per second). And comparing the bandwidth occupancy rates of all the links, so that whether the traffic load of the links is balanced can be judged. As an example, whether the traffic load is balanced may be determined according to whether the difference between the bandwidth occupancy rates of the links is greater than the threshold, and when the difference between the bandwidth occupancy rates of the links is greater than the threshold, the traffic load is considered to be unbalanced, and the setting of the threshold may be set according to the requirement. The bandwidth occupancy rate refers to the proportion of the data volume of the data stream forwarded by the link in unit time to the bandwidth of the link. For example, the critical value of the bandwidth occupancy rate difference of each link is set to 30%, the forwarding device may forward the data stream to the same client through two forwarding paths, link a and link B, where the bandwidths of link a and link B are both 10 Gbit/s. The data volume forwarded by the link A per second is 2Gbit/s, the data volume forwarded by the link A per second is 6Gbit/s, the bandwidth occupancy rate of the link A is 20%, the bandwidth occupancy rate of the link B is 60%, the difference value of the bandwidth occupancy rates of the two links is 40%, and is greater than the critical value 30%, so that the traffic loads of the link A and the link B are unbalanced.
S120, under the condition that the bandwidth occupancy rates of all the links are not balanced, identifying the types of data streams in the links of which the bandwidth occupancy rates exceed a threshold value, wherein the types of the data streams comprise elephant streams;
the reason for the imbalance of the link bandwidth occupancy rate is generally that a plurality of elephant flows and mouse flows exist in the data flow to be forwarded at the same time, and the plurality of elephant flows are forwarded through the same path, so that the data flow type in the overloaded link needs to be identified, and whether the elephant flow exists is judged. And if the elephant flow exists, switching the identified elephant flow to other unarmed paths for forwarding. Generally, even if traffic load is unbalanced under a light load condition, service experience is not affected, and in this case, a forwarding path of a data stream does not need to be switched, for example, the bandwidth occupancy rate of the link a is 10%, and the bandwidth occupancy rate of the link B is 30%. Therefore, a threshold needs to be set for the bandwidth occupancy of each link, and on the premise that the bandwidth occupancy is unbalanced, if the bandwidth occupancy of a link exceeds the threshold, the type of the data stream in the link is identified. The embodiments of the present description identify elephant flows.
In an embodiment of this specification, a data volume of each acquired data stream is obtained by sampling a data stream transmitted at an egress port of a forwarding device connected to a link whose bandwidth occupancy exceeds a threshold, and whether a elephant stream exists is determined according to the data volume of each data stream. Sampling can be performed by using different sampling techniques, for example, according to an sflow rule or a netstream rule, a specific sampling manner and an appropriate sampling rate are configured in the sflow rule or the netflow rule, and the collected samples are messages in each data stream. The forwarding device can perform periodic sampling by taking the number of messages as a unit, for example, the sampling rate is 1/1000 when the forwarding device acquires every 1000 messages; the forwarding device may also perform periodic sampling in units of time, such as once per second, with a sampling rate of 1/s. To prevent the impact on the CPU of the forwarding device, the sampling rate may be configured according to the processing power of the CPU, or configured according to the data amount of the data stream.
In an embodiment of the present description, the size of the data flow data is characterized by the ratio of the number of the collected packets carrying the same quintuple information in each data flow. Each collected message carries a group of quintuple information, and the messages carrying the same quintuple information belong to the same data stream. The number of the messages carrying the same quintuple information is equivalent to the data volume of the data flow corresponding to the quintuple information, and the larger the number of the messages is, the larger the data volume of the data flow is. And acquiring the proportion of the number of the messages carrying the same quintuple information in all the collected messages, and determining the data stream corresponding to the quintuple information as the elephant stream if the proportion reaches a preset value. For example, the forwarding device collects 100 messages, wherein 70 messages carry the same five-tuple information a, 20 messages carry the same five-tuple information B, and 10 messages carry the same five-tuple information C; and setting the data flow corresponding to the quintuple information with the message quantity ratio of the same quintuple information exceeding 60% as the elephant flow, and determining the data flow corresponding to the quintuple information A as the elephant flow. The preset value of the proportion is set according to the requirements of users and the traffic.
Another embodiment of the present specification represents the size of the data volume of each acquired data flow jointly by the transmission rate of each data flow and the ratio of the number of packets carrying the same quintuple information. The transmission rate of each data stream is calculated according to traffic statistics information of each data stream at an output port or an input port of the forwarding device, where the traffic statistics information includes a data volume transmitted by each data stream in a period of time. For example, the packets collected by sampling belong to three different data streams A, B and C, and their corresponding quintuple information is a, b and C, respectively. And issuing flow statistical instructions aiming at the quintuple information a, b and C at an output port of the forwarding equipment, and assuming that the data volume transmitted by the obtained data streams A, B and C in 1 second is respectively 6Gbit, 2Gbit and 1Gbit, the transmission rates of the data streams A, B and C are respectively 6Gbit/s, 2Gbit/s and 1 Gbit/s. And if the transmission rate of the data stream corresponding to the quintuple information exceeds a preset upper limit while the number proportion of the messages carrying the same quintuple information exceeds a preset value, determining the data stream corresponding to the quintuple information as the elephant stream. The preset upper limit of the transmission rate is set by the user himself. In addition, statistical resources used when performing traffic statistics need to be recovered. The flow statistic instruction issued during the flow statistic can be regarded as a counter, the counter is bound with quintuple information of the data flow counted by the counter, the bound counter cannot execute other statistic tasks, and the number of the counters is limited. Therefore, in order to reasonably use the statistical resources, the issued traffic statistical instruction needs to be timely recycled.
S130, switching the identified elephant flow to a light-load link according to an ACL rule, wherein the ACL rule is configured according to the identified quintuple information of the elephant flow, and the light-load link is a link of which the bandwidth occupancy rate does not exceed the threshold value.
At this time, switching the forwarding path of the identified elephant flow to another link whose bandwidth occupancy rate does not exceed the threshold value requires issuing a rule, ACL (Access Control List), whose priority is higher than that of the routing table. The ACL is a set of one or more rules, where the rules refer to judgment statements describing message matching conditions, and the conditions may be a source address, a destination address, a port number, and the like of the message. The device performs message matching based on these rules, can filter out specific messages, and allow or prevent the messages from passing through according to the processing policy of the service module applying the ACL. One embodiment of the present description enables identified elephant flows to be forwarded from links whose bandwidth occupancy does not exceed a threshold by employing a PBR (Policy Based Routing) configured with ACL rules. Firstly, an administrator or an SDN controller establishes an ACL rule according to the identified quintuple information of the elephant flow, allows the elephant flow to pass through a forwarding device, then configures a PBR calling the ACL rule, and sets a next hop address of the elephant flow forwarded from a light load link in the PBR. Since the priority of the PBR is higher than the forwarding path result calculated by the ECMP through the HASH, the forwarding device applies the PBR issued by the administrator or the SDN controller to switch the forwarding path of the identified elephant flow to a light-load link that does not exceed the threshold.
A method for implementing ECMP traffic load balancing according to the present application is described below with reference to a preferred embodiment, and fig. 2 is a schematic view of an application scenario. The data flows f1, f2, f3 and f4 are all forwarded to the same client 40 by the forwarding device 10, where f1 and f3 are mouse flows and f2 and f4 are elephant flows. There are two shortest forwarding PATHs, the link where the device 20 is located is PATH1, the link where the device 30 is located is PATH2, and the bandwidths of PATH1 and PATH2 are both 10 Gbit/s. After HASH calculation, f1 and f3 are forwarded via PATH1, and f2 and f4 are forwarded via PATH 2.
As shown in fig. 3, a method for implementing ECMP traffic load balancing proposed by the present application includes the following steps:
s210, monitoring the bandwidth occupancy rates of two forwarding links, and judging whether the bandwidth occupancy rates of all the links are balanced;
assuming that after the forwarding PATHs f1, f1, f2, f3 and f4 are determined through HASH calculation, the bandwidth occupancy rate of the PATH1 is 10% and the bandwidth occupancy rate of the PATH2 is 60%, the bandwidth occupancy rates are unbalanced.
S220, under the condition that the bandwidth occupancy rates of all links are not balanced, sampling data streams transmitted by the links of which the bandwidth occupancy rates exceed a threshold value at an output port of the forwarding equipment according to an sflow rule, wherein the sampling rate is 1/1000;
if the preset threshold of the bandwidth occupancy rate is 50%, the bandwidth occupancy rate of the PATH2 exceeds the threshold, the data streams f2 and f4 transmitted by the PATH2 are subjected to message sampling, and the sampling is performed once every 1000 messages are transmitted.
S230, acquiring the ratio of the number of the messages carrying the same quintuple information in the collected sample;
the messages carrying the same quintuple information belong to the same data flow. Suppose that 1000 messages are collected in total, wherein 700 messages carry a same group of quintuple information and belong to a data flow f 2; the other 300 messages carry the same quintuple information of another group, belong to the data flow f4, the message quantity ratio belonging to the data flow f2 is 70%, and the message quantity ratio belonging to the data flow f4 is 30%. The higher the duty ratio, the larger the data amount of the data stream, and the more likely it is to be a elephant stream.
S240, acquiring flow statistical information of each acquired data flow at an outlet port of the router, and calculating the transmission rate of each data flow;
and issuing a flow statistic instruction aiming at the five-tuple information of each acquired data flow to acquire the transmission rate of each data flow at the outlet port of the router. Suppose that the transmission rate of f2 is 4Gbit/s and the transmission rate of f4 is 2 Gbit/s. A higher transmission rate also indicates a larger amount of data for the data stream, which is more likely to be a elephant stream.
S250, determining the data flow with the message quantity ratio of the same quintuple information exceeding a preset value and the transmission rate exceeding a preset upper limit as a elephant flow;
the ratio preset value of the preset message quantity is 60%, the preset upper limit of the transmission rate is 3Gbit/s, and then the data flow f2 is determined to be the elephant flow.
And S260, switching the transmission path of the elephant flow into a light load link of which the bandwidth occupancy rate does not exceed a threshold value according to a PBR issued by an administrator or an SDN controller, wherein the PBR is configured with an ACL rule generated according to the five-tuple information of the identified elephant flow.
The data f2 is forwarded through PATH 1.
Based on the method performed by any of the above embodiments, the present application further provides a computer program product comprising a computer program which, when executed by a processor, performs the method performed by any of the above embodiments.
Embodiments of the computer program product of the present description may be applied to an electronic device, such as a forwarding device. Embodiments of the computer program product may be implemented by software, or by hardware, or by a combination of both. Taking the software implementation as an example, the software implementation is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for running through the processor in which the file processing is located. From a hardware aspect, as shown in fig. 4, which is a hardware structure diagram of an electronic device in which a computer program product is located in an embodiment of this specification, except for the processor 310, the memory 330, the network interface 320, and the nonvolatile memory 340 shown in fig. 4, an electronic device in which a computer program product 331 is located in an embodiment may also include other hardware according to an actual function of the computer device, and details of this are not described again.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Other embodiments of the present description will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This specification is intended to cover any variations, uses, or adaptations of the specification following, in general, the principles of the specification and including such departures from the present disclosure as come within known or customary practice within the art to which the specification pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the specification being indicated by the following claims.
It will be understood that the present description is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.
The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (10)
1. A method for realizing ECMP flow load balance is applied to forwarding equipment and is characterized by comprising the following steps:
monitoring the bandwidth occupancy rate of each link to which a plurality of data flows with the same target address are allocated, wherein the forwarding equipment allocates the links to the plurality of data flows with the same target address based on a Hash algorithm;
under the condition that the bandwidth occupancy rates of the links are not balanced, identifying the types of data streams in the links of which the bandwidth occupancy rates exceed a threshold value, wherein the types of the data streams comprise elephant streams;
and switching the identified elephant flow to a light-load link according to an ACL rule, wherein the ACL rule is configured according to the five-tuple information of the identified elephant flow, and the light-load link is a link of which the bandwidth occupancy rate does not exceed the threshold value.
2. The method of claim 1, wherein the identifying the type of data flow in the link having the bandwidth occupancy exceeding the threshold comprises:
sampling data streams transmitted by the links of which the bandwidth occupancy rates exceed the threshold value at the output port of the forwarding equipment according to a preset sampling rate;
and judging whether the elephant flow exists according to the data volume of each acquired data flow.
3. The method according to claim 2, wherein the data size of each collected data flow is characterized by the ratio of the number of the collected packets carrying the same quintuple information in each collected data flow.
4. The method of claim 3, wherein the determining whether an elephant flow exists according to the size of the data volume of each collected data stream comprises:
and if the number proportion of the messages carrying the same quintuple information exceeds a preset value, determining the data flow corresponding to the quintuple information as the elephant flow.
5. The method according to claim 3, wherein the size of the data volume of each collected data flow is further characterized by a transmission rate of the data flow, and the transmission rate is calculated according to traffic statistics information of each data flow at an ingress port or an egress port of a forwarding device.
6. The method of claim 5, wherein the determining whether the elephant flow exists according to the size of the collected data flow comprises:
and determining the data flow with the message quantity ratio carrying the same quintuple information exceeding a preset value and the transmission rate exceeding a preset upper limit as the elephant flow.
7. The method according to claim 2, wherein the sampling of the data stream transmitted via the link whose bandwidth occupancy rate exceeds the threshold at the egress port of the forwarding device according to the preset sampling rate is performed according to sflow or netstream rules.
8. The method of claim 1, wherein the switching the identified elephant flow to a light load link in accordance with ACL rules comprises:
receiving a PBR policy route issued by an administrator or an SDN controller, wherein the PBR is configured with an ACL rule generated according to the identified quintuple information of the elephant flow;
and replacing the transmission path of the elephant flow with a light load link according to the PBR.
9. A computer program product comprising a computer program, characterized in that the computer program realizes the method of any of claims 1-8 when executed by a processor.
10. An electronic device, characterized by comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111348567.4A CN114095441A (en) | 2021-11-15 | 2021-11-15 | Method for realizing ECMP flow load balance and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111348567.4A CN114095441A (en) | 2021-11-15 | 2021-11-15 | Method for realizing ECMP flow load balance and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114095441A true CN114095441A (en) | 2022-02-25 |
Family
ID=80300836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111348567.4A Withdrawn CN114095441A (en) | 2021-11-15 | 2021-11-15 | Method for realizing ECMP flow load balance and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114095441A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117579543A (en) * | 2024-01-16 | 2024-02-20 | 苏州元脑智能科技有限公司 | Data stream segmentation method, device, equipment and computer readable storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109218216A (en) * | 2017-06-29 | 2019-01-15 | 中兴通讯股份有限公司 | Chain polymerization traffic distribution method, device, equipment and storage medium |
| US20190327174A1 (en) * | 2016-07-01 | 2019-10-24 | Hewlett Packard Enterprise Development Lp | Load balancing |
-
2021
- 2021-11-15 CN CN202111348567.4A patent/CN114095441A/en not_active Withdrawn
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190327174A1 (en) * | 2016-07-01 | 2019-10-24 | Hewlett Packard Enterprise Development Lp | Load balancing |
| CN109218216A (en) * | 2017-06-29 | 2019-01-15 | 中兴通讯股份有限公司 | Chain polymerization traffic distribution method, device, equipment and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117579543A (en) * | 2024-01-16 | 2024-02-20 | 苏州元脑智能科技有限公司 | Data stream segmentation method, device, equipment and computer readable storage medium |
| CN117579543B (en) * | 2024-01-16 | 2024-04-05 | 苏州元脑智能科技有限公司 | Data stream segmentation method, device, equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Curtis et al. | Mahout: Low-overhead datacenter traffic management using end-host-based elephant detection | |
| US10193807B1 (en) | Penalty-box policers for network device control plane protection | |
| US6928482B1 (en) | Method and apparatus for scalable process flow load balancing of a multiplicity of parallel packet processors in a digital communication network | |
| US20210392043A1 (en) | Modifying resource allocation or policy responsive to control information from a virtual network function | |
| US8537846B2 (en) | Dynamic priority queue level assignment for a network flow | |
| US11595315B2 (en) | Quality of service in virtual service networks | |
| Trestian et al. | OFLoad: An OpenFlow-based dynamic load balancing strategy for datacenter networks | |
| Carpio et al. | DiffFlow: Differentiating short and long flows for load balancing in data center networks | |
| JP7313480B2 (en) | Congestion Avoidance in Slice-Based Networks | |
| Jeong et al. | Application-aware traffic engineering in software-defined network | |
| CN112954069A (en) | Method, device and system for accessing mobile equipment to SD-WAN (secure digital-Wide area network) | |
| US11102065B2 (en) | Detecting and resolving multicast traffic performance issues | |
| Wang et al. | Control link load balancing and low delay route deployment for software defined networks | |
| Hong et al. | An optimized flow management mechanism in OpenFlow network | |
| Duliński et al. | MPLS-based reduction of flow table entries in SDN switches supporting multipath transmission | |
| Tomovic et al. | A new approach to dynamic routing in SDN networks | |
| CN | A proactive flow admission and re-routing scheme for load balancing and mitigation of congestion propagation in SDN data plane | |
| Liu et al. | An SDN-based hybrid strategy for load balancing in data center networks | |
| Prakash et al. | Server-based dynamic load balancing | |
| Rashid | Sorted-GFF: An efficient large flows placing mechanism in software defined network datacenter | |
| Guo et al. | Oversubscription bounded multicast scheduling in fat-tree data center networks | |
| CN114095441A (en) | Method for realizing ECMP flow load balance and electronic equipment | |
| Hwang et al. | Load balancing and routing mechanism based on software defined network in data centers | |
| Ilvesmäki et al. | Flow classification schemes in traffic-based multilayer IP switching—comparison between conventional and neural approach | |
| Mauricio et al. | Aclflow: An nfv/sdn security framework for provisioning and managing access control lists |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220225 |
|
| WW01 | Invention patent application withdrawn after publication |