CN109450793B

CN109450793B - Method and device for scheduling service traffic

Info

Publication number: CN109450793B
Application number: CN201811547004.6A
Authority: CN
Inventors: 兰椿树
Original assignee: Ruijie Networks Co Ltd
Current assignee: Ruijie Networks Co Ltd
Priority date: 2018-12-18
Filing date: 2018-12-18
Publication date: 2021-07-16
Anticipated expiration: 2038-12-18
Also published as: CN109450793A

Abstract

The invention provides a method and a device for scheduling service traffic, comprising the following steps: establishing a tunnel, configuring the tunnel priority of the tunnel, and adding the identifiers of the tunnels with the same priority to the same ECMP group, wherein the priority of the tunnel is the priority of the ECMP group; establishing a corresponding relation among the service identification, the service priority and the tunnel priority; establishing a first ACL table item, wherein the first ACL table item comprises a corresponding relation between a service identifier and a service priority; establishing a second ACL table item, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority; searching a corresponding first service priority from a first ACL table item according to a first service message service identifier; searching a corresponding first tunnel priority from a second ACL table item according to the first service priority; determining an ECMP group corresponding to the first tunnel priority according to the first tunnel priority; and selecting a first tunnel from the ECMP group to send the first service message. The problem that a chip needs to identify different service message flows and needs to consume a large amount of forwarding table resources is solved.

Description

Method and device for scheduling service traffic

Technical Field

The present invention relates to the field of data communications, and in particular, to a method and an apparatus for scheduling traffic.

Background

In recent years, with the continuous development of internet technologies such as cloud computing and big data, the scale of a data center network is gradually enlarged, which puts higher requirements on the construction of the data center network of an ISP service provider, and how to effectively manage the network traffic of the data center, improve the network utilization rate and reduce the operation cost is a problem which needs to be solved urgently at present.

SR (segment routing) is a source routing-based network protocol with innovative significance as ethernet domain following multi-protocol label switching protocol MPLS. Compared with the traditional MPLS mechanism, the method has a simpler control plane and an easily-expanded data plane. By pressing the segment identifier in the source routing node, the message forwarding path can be effectively planned, and the flow control management can be well realized by combining the segment identifier with an SDN (Software Defined Network) controller.

In SRTE (Segment Routing Traffic Engineering) actual networking, there are many Traffic based on flow, IP address and application, and in order to realize effective control of Traffic, source Routing needs to prioritize them so as to provide differentiated services. For example: a data center metropolitan area network of a certain client adopts an SR network, priority differentiation can be carried out on service traffic on the requirement of a source node, meanwhile, traffic with different priorities can enter traffic engineering TE tunnels with different levels, and the traffic can be in load balance in the TE tunnels with the same level.

Pure IP traffic received from a CE (Customer Edge device) direction is matched with message characteristics (such as DSCP Differentiated Services Code Point) at an entry port of an operator Edge router PE, TE tunnels (TE tunnels) of 1, 2, 3 and 4 levels are respectively entered according to the mapping relation of DSCP- > TE levels, traffic with MPLS encapsulation received from a P (operator Provider) direction is matched with EXP at an entry port of P, and TE tunnels of 1, 2, 3 and 4 levels are respectively entered according to the mapping relation of EXP- > TE levels.

When the network switch equipment of the operator needs to prioritize and forward the service traffic, one method is to pass the policy Routing, and it is characterized in that the forwarding equipment can select the Routing according to the policy set by the user, the packet data message can directly appoint the exit (the exit may be a tunnel) without searching the Routing table, it is more flexible, the disadvantage of this method is that the method using manual configuration is easy to leak and the workload is larger, and does not support ECMP (Equal cost-Costmultipath Routing), and it needs to consume more ACL hardware resources.

Disclosure of Invention

In order to solve the technical problem, the embodiment of the invention adopts the following technical scheme:

a method for scheduling traffic flow comprises the following steps:

establishing a tunnel, configuring the tunnel priority of the tunnel, and adding the identifiers of the tunnels with the same priority to the same ECMP group, wherein the priority of the tunnel is the priority of the ECMP group;

establishing a corresponding relation among the service identifier, the service priority and the tunnel priority;

receiving a first service message, and searching a corresponding first service priority from a first ACL table item according to a service identifier of the first service message, wherein the first ACL table item comprises a corresponding relation between the service identifier and the service priority;

searching a corresponding first tunnel priority from a second ACL table item according to the first service priority, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority;

searching a forwarding table according to the IP address of the first service message, and determining a first ECMP group identifier corresponding to the first service message, wherein the forwarding table comprises a corresponding relation between the IP address and the ECMP group identifier;

determining a first ECMP group corresponding to the first tunnel priority according to the first ECMP group identifier and the first tunnel priority;

and selecting a first tunnel from the first ECMP group according to a load balancing strategy to send the first service message.

Optionally, the service identifier is a combination of one or more of a source/destination IP address, a protocol type, a source/destination port number, and a DSCP of the service packet.

Optionally, the number of the service priorities is the same as the number of the tunnel priorities.

Optionally, the tunnel is an SRTE tunnel.

Optionally, the forwarding table is a routing table or an MPLS forwarding table.

Another aspect of the embodiments of the present invention is to provide a device for scheduling service traffic, including:

the first establishing module is used for establishing a tunnel, configuring the tunnel priority of the tunnel, and adding the identifiers of the tunnels with the same priority to the same ECMP group, wherein the priority of the tunnel is the priority of the ECMP group;

the second establishing module is used for establishing the corresponding relation among the service identifier, the service priority and the tunnel priority;

a receiving module, configured to receive a first service packet,

a searching module, configured to search a corresponding first service priority from a first ACL entry according to a service identifier of the first service packet, where the first ACL entry includes a correspondence between the service identifier and the service priority, and search a corresponding first tunnel priority from a second ACL entry according to the first service priority, where the second ACL entry includes a correspondence between the service priority and the tunnel priority, search a forwarding table according to an IP address of the first service packet, determine a first ECMP group identifier corresponding to the first service packet, where the forwarding table includes a correspondence between the IP address and the ECMP group identifier;

a determining module, configured to determine, according to the first ECMP group identifier and the first tunnel priority, a first ECMP group corresponding to the first tunnel priority;

and the sending module is used for selecting a first tunnel from the first ECMP group according to a load balancing strategy to send the first service message.

Optionally, the tunnel is an SRTE tunnel.

The embodiment of the invention has the advantages that a method for scheduling the service flow is provided, a tunnel is established, the priority of the tunnel is configured, the identifiers of the tunnels with the same priority are added into the same ECMP group, and the priority of the tunnel is the priority of the ECMP group; establishing a corresponding relation among the service identifier, the service priority and the tunnel priority; receiving a first service message, and searching a corresponding first service priority from a first ACL table item according to a service identifier of the first service message, wherein the first ACL table item comprises a corresponding relation between the service identifier and the service priority; searching a corresponding first tunnel priority from a second ACL table item according to the first service priority, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority; searching a forwarding table according to the IP address of the first service message, and determining a first ECMP group identifier corresponding to the first service message, wherein the forwarding table comprises a corresponding relation between the IP address and the ECMP group identifier; determining an ECMP group corresponding to the first tunnel priority according to the first ECMP group identifier and the first tunnel priority; and selecting a first tunnel from the ECMP group according to a load balancing strategy to send the first service message. Through the two-stage ACL lookup function, the message outlet is selected according to the priority of the service message flow on the basis of the traditional routing lookup result, and the consumption of hardware resources can be greatly reduced. On the basis of not increasing network cost, the problem that a switching chip needs to identify different service message flows and needs to consume a large amount of forwarding table resources is solved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a flow chart of a method according to an embodiment of the present invention;

FIG. 2 is a diagram of an apparatus according to an embodiment of the present invention;

FIG. 3 is a diagram illustrating a priority policy matching according to an embodiment of the present invention;

fig. 4 is a network topology structure diagram according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

A method for scheduling traffic, as shown in fig. 1, includes:

s101, establishing a tunnel, configuring the tunnel priority of the tunnel, and adding the identifiers of the tunnels with the same priority to the same ECMP group, wherein the priority of the tunnel is the priority of the ECMP group;

s103, establishing a corresponding relation among the service identifier, the service priority and the tunnel priority;

s105, receiving a first service message, and searching a corresponding first service priority from a first ACL table item according to a service identifier of the first service message, wherein the first ACL table item comprises a corresponding relation between the service identifier and the service priority;

s107, searching a corresponding first tunnel priority from a second ACL table item according to the first service priority, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority;

s109, searching a forwarding table according to the IP address of the first service message, and determining a first ECMP group identifier corresponding to the first service message, wherein the forwarding table comprises a corresponding relation between the IP address and the ECMP group identifier;

s111, determining a first ECMP group corresponding to the first tunnel priority according to the first ECMP group identifier and the first tunnel priority;

s113, selecting a first tunnel from the first ECMP group according to a load balancing policy, and sending the first service packet.

Optionally, the tunnel is an SRTE tunnel.

The embodiment of the invention has the advantages that a method for scheduling the service flow is provided, a tunnel is established, the priority of the tunnel is configured, the identifiers of the tunnels with the same priority are added into the same ECMP group, and the priority of the tunnel is the priority of the ECMP group; establishing a corresponding relation among the service identifier, the service priority and the tunnel priority; receiving a first service message, and searching a corresponding first service priority from a first ACL table item according to a service identifier of the first service message, wherein the first ACL table item comprises a corresponding relation between the service identifier and the service priority; searching a corresponding first tunnel priority from a second ACL table item according to the first service priority, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority; searching a forwarding table according to the IP address of the first service message, and determining a first ECMP group identifier corresponding to the first service message, wherein the forwarding table comprises a corresponding relation between the IP address and the ECMP group identifier; determining a first ECMP group corresponding to the first tunnel priority according to the first ECMP group identifier and the first tunnel priority; and selecting a first tunnel from the first ECMP group according to a load balancing strategy to send the first service message. Through the two-stage ACL lookup function, the message outlet is selected according to the priority of the service message flow on the basis of the traditional routing lookup result, and the consumption of hardware resources can be greatly reduced. On the basis of not increasing network cost, the problem that a switching chip needs to identify different service message flows and needs to consume a large amount of forwarding table resources is solved

Another aspect of the embodiments of the present invention is to provide an apparatus for scheduling service traffic, as shown in fig. 2, including:

a first establishing module 201, configured to establish a tunnel, configure tunnel priorities of the tunnels, and add identifiers of tunnels with the same priority to a same ECMP group, where the priority of the tunnel is the priority of the ECMP group;

a second establishing module 203, configured to establish a correspondence between the service identifier, the service priority, and the tunnel priority;

a receiving module 205, configured to receive a first service packet,

a searching module 207, configured to search a corresponding first service priority from a first ACL entry according to a service identifier of the first service packet, where the first ACL entry includes a correspondence between the service identifier and the service priority, and search a corresponding first tunnel priority from a second ACL entry according to the first service priority, where the second ACL entry includes a correspondence between the service priority and the tunnel priority, search a forwarding table according to an IP address of the first service packet, determine a first ECMP group identifier corresponding to the first service packet, where the forwarding table includes a correspondence between an IP address and an ECMP group identifier;

a determining module 209, configured to determine, according to the first ECMP group identifier and the first tunnel priority, a first ECMP group corresponding to the first tunnel priority;

a sending module 211, configured to select a first tunnel from the first ECMP group according to a load balancing policy to send the first service packet.

Optionally, the tunnel is an SRTE tunnel.

The embodiment of the invention has the advantages that a method for scheduling the service flow is provided, a tunnel is established, the priority of the tunnel is configured, the identifiers of the tunnels with the same priority are added into the same ECMP group, and the priority of the tunnel is the priority of the ECMP group; establishing a corresponding relation among the service identifier, the service priority and the tunnel priority; receiving a first service message, and searching a corresponding first service priority from a first ACL table item according to a service identifier of the first service message, wherein the first ACL table item comprises a corresponding relation between the service identifier and the service priority; searching a corresponding first tunnel priority from a second ACL table item according to the first service priority, wherein the second ACL table item comprises the corresponding relation between the service priority and the tunnel priority; searching a forwarding table according to the IP address of the first service message, and determining a first ECMP group identifier corresponding to the first service message, wherein the forwarding table comprises a corresponding relation between the IP address and the ECMP group identifier; determining a first ECMP group corresponding to the first tunnel priority according to the first ECMP group identifier and the first tunnel priority; and selecting a first tunnel from the first ECMP group according to a load balancing strategy to send the first service message. Through the two-stage ACL lookup function, the message outlet is selected according to the priority of the service message flow on the basis of the traditional routing lookup result, and the consumption of hardware resources can be greatly reduced. On the basis of not increasing network cost, the problem that a switching chip needs to identify different service message flows and needs to consume a large amount of forwarding table resources is solved.

The following further describes the embodiments of the present invention with reference to a specific application scenario of the SRTE tunnel, but the description should not be construed as limiting the scope of the present invention:

in the application scenario, service priority mapping may be configured on a core switch, and the steps are as follows:

step one, according to the IP address, protocol type, port number, DSCP and other network characteristic elements of the service flow (service message), the priority of the service is distinguished. For example, a Class Service may be identified by "DSCP" or "protocol type" alone, or a Class Service may be identified by "IP + DSCP" or "IP + protocol port number", the Service priority is represented by Service Class, and the number of Service priorities may be determined by client configuration.

And step two, configuring the priority attribute of the SRTE tunnel, wherein the priority attribute corresponds to a Service Class. For example, DSCP is specified here as a service class identity:

service 1: when DSCP is 20, Service Class is 2;

service 2: when DSCP is 30, Service Class is 3;

the service 1 and the service 2 are required to be forwarded in the SRTE tunnels with the priorities of 2 and 3, respectively, the specific mapping relationship is shown in table 1, and the corresponding relationship among the service identifier, the service priority and the tunnel priority is established.

Business	Identification (DSCP)	Priority of service	Priority of tunnel
				Service
1	20	2	2
				Service 2	30	3	3

TABLE 1

After the above configuration is made, the switch hardware (chip) implementation flow is as follows, but in other embodiments, there is no precedence relationship between the above configuration process and the switch hardware implementation flow:

step one, establishing an SRTE tunnel, configuring the priority of the tunnel (as shown in table 1, may be the same as the Service priority Class), and placing the tunnel with the same priority in the same ECMP group, such as the ECMP group (ECMP0, ECMP1, ECMP2, ECMP3, where 0, 1, 2, and 3 are not the ID of the ECMP, but only represent the first priority ECMP group, the second priority ECMP group, the third priority ECMP group, and the fourth priority ECMP group, respectively) in fig. 3, and respectively correspond to the traffic flow (assuming that the traffic flow with 4 priorities is planned) whose Service priority (Service Class) is (1, 2, 3, 4).

Step two, a forwarding table (e.g., a routing table) is established, the ethernet IP service packet enters the switch, and the forwarding table contains a corresponding relationship between an IP address and an ECMP group identifier, and then the forwarding table can be searched according to a destination IP address DIP to obtain an ECMP group identifier ID (e.g., the ID of ECMP0 is 0x0010, the ID of ECMP1 is 0x0011, the ID of ECMP2 is 0x0012, and the ID of ECMP3 is 0x 0013).

And step three, establishing a second-level ACL table item, wherein an ACL matching domain is a service priority value SCValue, only one ACL table item can be installed in the second-level ACL table item, and when a service message is processed subsequently, the chip can automatically add the ECMPID searched in the step two to the service priority value SCValue obtained in ACL matching to obtain a new ECMPID, wherein the new ECMP ID is pointed to an ECMP group which is used for storing the SRTE tunnel with priority in advance.

If the ID of the ECMP0 found in step two is 0x0010, and the value SCValue of the service priority obtained from the ACL matching is 2, the ECMP 0+2 is 0x0010+2 is 0x0012, 0x0012 is a new ECMP ID, and the new ECMP ID is the ID of ECMP2, then ECMP2 points to the ECMP group that is used to store the SRTE tunnel with priority in advance.

And step four, establishing a first-level ACL table item for matching the characteristics of the Service messages (can be in the header information of the Service messages), dividing the Service messages with the same characteristics into a priority group (Service Class), and using SCValue on a chip to express the priority of the packet messages.

The process of processing the service message is as follows:

step one, receiving a service message 1, and searching a corresponding service priority 2 from a first ACL table item (namely, a first-level ACL table item) according to a service identifier 20 (which can be in header information of the service message) of the service message 1;

step two, searching a forwarding table according to the IP address of the service message 1 to obtain an ECMP group identifier ID of 0x0010, searching a corresponding tunnel priority 2 from the second ACL entry (i.e., the second ACL entry) according to the service priority 2, or adding the service priority 2 to the ID value 0x0010 of the ECMP0 to be equal to 0x0012, and obtaining a new ECMP group ECMP2 according to 0x 0012;

step three, determining that an ECMP group corresponding to the tunnel priority is ECMP2 according to the tunnel priority 2 and the ECMP 0; the ECMP2 or the new ECMP group ECMP2 in step two points to the ECMP group that was previously used to deposit the SRTE tunnel with priority.

And step four, selecting a first tunnel from the ECMP2 group according to a load balancing strategy to send the first service message.

As shown in fig. 4, the virtual private network L3VPN and SRTE hybrid networking diagram is a simple third-layer-based virtual private network L3VPN and SRTE hybrid networking diagram, in this scenario, CE1 and CE2 access an SRTE network through PE1 and PE2, respectively, there are 4 groups of SRTE tunnels (indicated by four solid lines between PE1 and PE 2) with different priorities in the SRTE network and form an ECMP, which is used for transmitting multiple traffic flows between CE1 and CE2, and if no traffic flow is prioritized, traffic congestion is easily caused as the traffic flows increase, so that various traffic flows are affected. By adopting the scheme described in the embodiment of the invention, the automatic routing of the service traffic can be easily realized as long as the embodiment of the invention containing two levels of ACLs is deployed on the nodes PE1 and PE 2.

The embodiment of the invention has the advantages that the two-stage ACL lookup function is adopted to realize the selection of the message outlet according to the priority of the service message flow on the basis of the traditional routing lookup result, thereby greatly reducing the consumption of hardware resources. On the basis of not increasing network cost, the problem that a switching chip needs to identify different service message flows and needs to consume a large amount of forwarding table resources is solved.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A method for scheduling traffic, comprising:

2. The method of claim 1, wherein the traffic identification is a combination of one or more of a source/destination IP address, a protocol type, a source/destination port number, and a DSCP of a traffic packet.

3. A method according to any of claims 1-2, characterized in that the number of traffic priorities is the same as the number of tunnel priorities.

4. The method of any of claims 1-2, wherein the tunnel is an SRTE tunnel.

5. The method of claim 1, wherein the forwarding table is a routing table or an MPLS forwarding table.

6. An apparatus for traffic scheduling, comprising:

a receiving module, configured to receive a first service packet,

7. The apparatus of claim 6, wherein the traffic identification is a combination of one or more of a source/destination IP address, a protocol type, a source/destination port number, and a DSCP of a traffic packet.

8. An apparatus according to any of claims 6-7, wherein the number of traffic priorities is the same as the number of tunnel priorities.

9. The apparatus of any of claims 6-7, wherein the tunnel is an SRTE tunnel.

10. The apparatus of claim 6, wherein the forwarding table is a routing table or an MPLS forwarding table.