Method for realizing continuous flow acquisition based on resource pool
By creating a policy matching table and forwarding policy table on the network processing chip of the shunt, the full forwarding and percentage polling forwarding are adopted to solve the problem of waste of traffic resources in the 5G network, and the continuous acquisition and light traffic is achieved Quantitative management reduces the number of server configurations and operation and maintenance costs.
Patent Information
- Application Number
- CN202310711847.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-15
- Publication Date
- 2024-12-17
AI Technical Summary
In the 5G era, the growth of massive Internet data has led to waste of resources. It is difficult for existing technologies to achieve real-time control of network operations and effective assessment of traffic while ensuring the normal operation of the network, especially because abnormal traffic accounts for a small proportion but requires A large number of resources are stored and analyzed, resulting in incalculable resource losses.
By creating a policy matching table and forwarding policy table on the network processing chip of the shunt, using full forwarding and percentage polling forwarding, the number of back-end server configurations is reduced and the continuous acquisition of traffic is achieved. The specific steps include creating a policy matching Tables and forwarding policy tables, new forwarding policy table action record bits and hash value ranges are added, and the table is queryed according to traffic characteristics and forwarding actions are performed.
It significantly reduces the number of back-end server resources, reduces operation and maintenance pressure and costs, realizes lightweight management and traffic evaluation, and improves the feasibility and progress of network construction plans.
Smart Images

Figure CN119155201A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to network communication technology, and in particular to a method for realizing continuous flow acquisition based on a resource pool. Background Art
[0002] The 5G era has arrived. Compared with 4G technology, 5G networks have stronger performance. However, with the technological innovations of higher data transmission rates, lower forwarding delays, and more connected applications, a massive amount of Internet data will grow.
[0003] The growth of Internet data also brings greater management requirements, but the vast majority of network traffic is safe and reliable, and the proportion of abnormal traffic is relatively small. However, if the entire traffic is directly stored and analyzed because of the need to evaluate the traffic of this small part, the resulting resource loss is immeasurable, and a large amount of manpower and material resources need to be invested in the early stage of the solution construction, which is not conducive to the gradual improvement and batch construction of the solution. Therefore, there is an urgent need for a method to save resources, while ensuring the normal operation of the network, to complete the real-time control of the actual operation of the network. To solve this problem, how to achieve continuous traffic acquisition based on the resource pool is a difficult point. Summary of the invention
[0004] In order to solve the above problems, the present invention proposes a method for realizing continuous acquisition of traffic based on a resource pool based on a diverter. Continuous acquisition of traffic is realized based on a resource pool, which greatly saves resources while ensuring functional completeness. The diverter is mainly used for multi-interface collection and policy-based diversion of network traffic. Network traffic is accessed through the hardware interface chip of the diverter. After the network traffic reaches the network processor from the interface chip, policy matching is performed. The network processor realizes refined diversion of traffic according to the matching results. However, the access processing capability of the diverter is relatively strong. A single 1U-height diverter can realize traffic access of more than 1Tbps, but the access processing capability of the server generally does not exceed 100Gbps. Therefore, when a single diverter has full traffic access, if continuous acquisition of traffic is required, hundreds of servers must be configured, which is not feasible for traffic construction solutions under certain network projects. This method is a functional innovation on the diverter network processing chip, which can greatly reduce the number of back-end server configurations under continuous traffic acquisition. The specific steps are as follows: Step 1: Create a policy matching table and a forwarding policy table on the network processor. The size of the policy matching table entry is 100*1000. The key of the policy matching table consists of five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information, and vlan / mpls label number. The five-tuple information occupies 13 bytes, the flow flag information occupies 1 byte, and the vlan / mpls label number information occupies 2 bytes. The result of the policy matching table is the forwarding policy table number; the size of the forwarding policy table entry is 32*1024. The key of the forwarding policy table consists of the forwarding policy table number (range 0-63) and the hash value (range 0-1023). The forwarding policy matching table number occupies the upper 6 bits and the hash value occupies the lower 10 bits. The result of the forwarding policy table consists of the forwarding target port number information and some information that needs to be carried.
[0005] Step 2: Add 32 forwarding policy table action record bits and 32 hash value ranges (maximum and minimum values) corresponding to the forwarding policy table on the network processor. The forwarding policy table can set two forwarding actions: full forwarding and percentage-based polling forwarding. Setting full forwarding means that all traffic that hits the policy will be forwarded from this policy forwarding group after the hit rate is recorded, without additional settings. When the forwarding policy table sets the forwarding action in proportion, for friendly interaction requirements, the customer only needs to set a percentage and polling time. For example, if the forwarding policy table sets 10% forwarding and the polling time is 10s, the control plane software will automatically write the hash value range corresponding to the forwarding policy match table to 0-101 at the beginning of the first start time, and change the hash value range corresponding to the forwarding policy match table to 102-203 at the beginning of the second start time. By analogy, the continuous acquisition of the full range of hashes of all traffic can be completed within 10 polling times. When the hash of the traffic is within the hash range of the forwarding policy table, the hit rate information is recorded and forwarded, otherwise only the hit rate information is recorded.
[0006] Step 3: When traffic is transferred from the hardware interface chip of the splitter to the network processor on the splitter, the network processor extracts the five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information and vlan / mpls label number according to the real-time characteristics of the traffic to form a 16-byte key query policy matching table; if the traffic does not find the policy matching table, the default action of the whole machine (usually set to discard) is executed to end the process; if the policy matching table is hit, jump to step 4.
[0007] Step 4: When traffic hits the policy match table, the forwarding policy table number is obtained according to the result information of the policy match table. The HASH value of the traffic is calculated according to the flow information. The method of calculating the HASH value is: source IP*1+destination IP*2+source port*3+destination port*4+protocol number*5. The remainder of the result to 1024 is the HASH value of the flow. The forwarding policy table number and the HASH value of the traffic form a key to query the forwarding policy table. At this time, if the action corresponding to the forwarding policy table is full forwarding, the traffic is forwarded directly according to the result of the forwarding policy table after counting the hit information; if the action corresponding to the forwarding policy table is polling forwarding, check whether the HASH of the traffic is within the hash range of the forwarding policy table. If it is within the range, the traffic is forwarded directly according to the result of the forwarding policy table after counting the hit information. Otherwise, only the hit information is counted and the traffic is discarded; the innovative method of forwarding by percentage polling can achieve continuous acquisition of traffic while saving resources.
[0008] The technical solution of the present invention above realizes continuous acquisition of traffic based on a resource pool. While ensuring functional completeness, compared with the prior art, its significant beneficial effects are: 1. Reduce the number of back-end server resources: At the early stage of the construction plan, the feasibility evaluation of the overall plan can be achieved at a lower cost.
[0009] 2. Lightweight management is achieved: Due to the reduction of the overall equipment of the solution, the pressure on later operation and maintenance is significantly reduced, reducing the operation and maintenance costs.
[0010] 3. Subsequent function iteration: After continuous traffic acquisition, it is convenient for customers to evaluate the overall traffic sample according to the actual traffic ratio and improve the progress of the solution construction. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, in which: Figure 1 It is a data forwarding plane flow chart of a method for realizing continuous traffic acquisition based on a resource pool provided by an embodiment of the present invention.
[0012] Figure 2 It is a control plane block diagram of a method for realizing continuous traffic acquisition based on a resource pool provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0013] In order to make the purpose and technical solution of the present disclosure more clearly understood, the present disclosure is further elaborated in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0014] like Figure 1 , Figure 2 As shown, an embodiment of a method for realizing continuous traffic acquisition based on a resource pool of the present invention includes creating a policy matching table and a forwarding policy table on a network processor, adding 32 forwarding policy table action record bits and 32 hash value ranges corresponding to the forwarding policy table, querying whether the traffic hits the policy matching table, and if it hits, obtaining the forwarding policy table number according to the result information of the policy matching table, and if it does not hit, executing the default action of the entire machine. The present invention has functional innovations on the splitter network processing chip, which can greatly reduce the number of back-end server configurations under continuous traffic acquisition. Realizing continuous traffic acquisition based on a resource pool greatly saves resources while ensuring functional completeness, reduces the number of server resources supporting the back-end, and realizes lightweight management and subsequent functional iteration.
[0015] The specific steps are as follows Step 1: Create a policy matching table and a forwarding policy table on the network processor. The size of the policy matching table entry is 100*1000. The key of the policy matching table consists of five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information, and vlan / mpls label number. The five-tuple information occupies 13 bytes, the flow flag information occupies 1 byte, and the vlan / mpls label number information occupies 2 bytes. The result of the policy matching table is the forwarding policy table number; the size of the forwarding policy table entry is 32*1024. The key of the forwarding policy table consists of the forwarding policy table number (range 0-63) and the hash value (range 0-1023). The forwarding policy matching table number occupies the upper 6 bits and the hash value occupies the lower 10 bits. The result of the forwarding policy table consists of the forwarding target port number information and some information that needs to be carried.
[0016] Step 2: Add 32 forwarding policy table action record bits and 32 hash value ranges (maximum and minimum values) corresponding to the forwarding policy table on the network processor. The forwarding policy table can set two forwarding actions: full forwarding and percentage-based polling forwarding. Setting full forwarding means that all traffic that hits the policy will be forwarded from this policy forwarding group after the hit rate is recorded, without additional settings. When the forwarding policy table sets the forwarding action in proportion, for friendly interaction requirements, the customer only needs to set a percentage and polling time. For example, if the forwarding policy table sets 10% forwarding and the polling time is 10s, the control plane software will automatically write the hash value range corresponding to the forwarding policy match table to 0-101 at the beginning of the first start time, and change the hash value range corresponding to the forwarding policy match table to 102-203 at the beginning of the second start time. By analogy, the continuous acquisition of the full range of hashes of all traffic can be completed within 10 polling times. When the hash of the traffic is within the hash range of the forwarding policy table, the hit rate information is recorded and forwarded, otherwise only the hit rate information is recorded.
[0017] Step 3: When traffic is transferred from the hardware interface chip of the splitter to the network processor on the splitter, the network processor extracts the five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information and vlan / mpls label number according to the real-time characteristics of the traffic to form a 16-byte key query policy matching table; if the traffic does not find the policy matching table, the default action of the whole machine (usually set to discard) is executed to end the process; if the policy matching table is hit, jump to step 4.
[0018] Step 4: When traffic hits the policy match table, the forwarding policy table number is obtained according to the result information of the policy match table. The HASH value of the traffic is calculated according to the flow information. The method of calculating the HASH value is: source IP*1+destination IP*2+source port*3+destination port*4+protocol number*5. The remainder of the result to 1024 is the HASH value of the flow. The forwarding policy table number and the HASH value of the traffic form a key to query the forwarding policy table. At this time, if the action corresponding to the forwarding policy table is full forwarding, the traffic is forwarded directly according to the result of the forwarding policy table after counting the hit information; if the action corresponding to the forwarding policy table is polling forwarding, check whether the HASH of the traffic is within the hash range of the forwarding policy table. If it is within the range, the traffic is forwarded directly according to the result of the forwarding policy table after counting the hit information. Otherwise, only the hit information is counted and the traffic is discarded; the innovative method of forwarding by percentage polling can achieve continuous acquisition of traffic while saving resources.
[0019] Various aspects of the present invention are described in this disclosure with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the present disclosure are not necessarily defined to include all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed by the present invention are not limited to any implementation. In addition, some aspects disclosed by the present invention can be used alone, or in any appropriate combination with other aspects disclosed by the present invention.
[0020] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. A person with ordinary knowledge in the technical field to which the present invention belongs may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the definition of the claims.
Claims
1. A method for achieving continuous traffic acquisition based on a resource pool, characterized in that: The method comprises: Step 1: Create a policy matching table and a forwarding policy table on the network processor. The size of the policy matching table entry is 100*1000. The key of the policy matching table consists of five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information and vlan / mpls label number. The five-tuple information occupies 13 bytes, the flow flag information occupies 1 byte, and the vlan / mpls label number information occupies 2 bytes. The result of the policy matching table is the forwarding policy table number; the size of the forwarding policy table entry is 32*1024. The key of the forwarding policy table consists of the forwarding policy table number (range 0-63) and the hash value (range 0-1023). The forwarding policy matching table number occupies the upper 6 bits and the hash value occupies the lower 10 bits. The result of the forwarding policy table consists of the forwarding target port number information and some information that needs to be carried. Step 2: Add 32 forwarding policy table action record bits and 32 hash value ranges (maximum and minimum values) corresponding to the forwarding policy tables on the network processor. The forwarding policy table can set two forwarding actions: full forwarding and percentage-based polling forwarding. Setting full forwarding means that all traffic that hits the policy records the hit rate and all are forwarded from the forwarding group of this policy without additional settings. When the forwarding policy table sets the forwarding action in proportion, for friendly interaction requirements, the customer only needs to set a percentage and polling time. For example, if the forwarding policy table sets 10% forwarding and the polling time is 10s, the control plane software will automatically write the hash value range corresponding to the forwarding policy match table to 0-101 at the beginning of the first start time, and change the hash value range corresponding to the forwarding policy match table to 102-203 at the beginning of the second start time. By analogy, the continuous acquisition of the full range of hashes of all traffic can be completed within 10 polling times. When the hash of the traffic is within the hash range of the forwarding policy table, the hit rate information is recorded and forwarded, otherwise only the hit rate information is recorded; Step 3: When traffic is transferred from the hardware interface chip of the splitter to the network processor on the splitter, the network processor extracts the five-tuple information (source IP, destination IP, source port, destination port, protocol number), flow flag information and vlan / mpls label number according to the real-time characteristics of the traffic to form a 16-byte key query policy matching table; if the traffic does not match the policy matching table, the default action of the whole machine (usually set to discard) is executed to end the process; if the traffic matches the policy matching table, jump to step 4; Step 4: When traffic hits the policy matching table, the forwarding policy table number is obtained according to the result information of the policy matching table, and the HASH value of the traffic is calculated according to the flow information. The method for calculating the HASH value is: source IP*1+destination IP*2+source port*3+destination port*4+protocol number*5. The remainder of the result to 1024 is the HASH value of the flow. The forwarding policy table number and the HASH value of the traffic form a key to query the forwarding policy table. At this time, if the action corresponding to the forwarding policy table is full forwarding, the traffic is forwarded directly according to the result of the forwarding policy table after the hit information is counted; if the action corresponding to the forwarding policy table is polling forwarding, check whether the HASH of the traffic is within the hash range of the forwarding policy table. If it is within the range, the hit information is counted and the traffic is forwarded directly according to the result of the forwarding policy table. Otherwise, only the hit information is counted and the traffic is discarded; the innovative method of forwarding by percentage polling can achieve continuous acquisition of traffic while saving resources.
