CN116723143B - Network target range resource allocation method and system based on traffic affinity - Google Patents
Network target range resource allocation method and system based on traffic affinity Download PDFInfo
- Publication number
- CN116723143B CN116723143B CN202311006748.8A CN202311006748A CN116723143B CN 116723143 B CN116723143 B CN 116723143B CN 202311006748 A CN202311006748 A CN 202311006748A CN 116723143 B CN116723143 B CN 116723143B
- Authority
- CN
- China
- Prior art keywords
- topology
- segment
- node
- virtualization
- list
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013468 resource allocation Methods 0.000 title claims abstract description 23
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000012163 sequencing technique Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000015654 memory Effects 0.000 claims description 23
- 230000009191 jumping Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 230000003993 interaction Effects 0.000 abstract description 4
- 238000012549 training Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/38—Flow based routing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/40—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using virtualisation of network functions or resources, e.g. SDN or NFV entities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/76—Routing in software-defined topologies, e.g. routing between virtual machines
-
- 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
Abstract
The invention discloses a network target range resource allocation method and system based on flow affinity, wherein the method firstly analyzes the target range scene topology to generate a directed graph based on a flow generator; calculating the flow size and the virtualization cost of each topological segment in the graph, and sequencing the topological segments and the calculation nodes respectively; then, virtualization of the directed graph is executed, and the node with the highest flow correlation is constrained to the same computing node; finally, the remaining non-traffic affinity nodes are instantiated again. The invention can reduce the cross-physical node interaction of the flow, reduce the risk of network storm, and greatly relieve the pressure of the exchanger for forwarding the flow. And further for the non-traffic affinity nodes, the non-traffic affinity nodes are started on the computing node with the most abundant current physical resources in each round of allocation process, so that the load balance of the total computing clusters can be ensured.
Description
Technical Field
The invention relates to a network target range resource allocation method and system based on traffic affinity, and belongs to the technical fields of virtualization technology and network security.
Background
The network target range is a test platform which simulates a real network space attack and defense combat environment through a virtualization technology and can support combat capability research and weapon equipment verification. In order to achieve a better training effect, interference flow is applied to the nodes in the actual use process. In planning training scenarios, traffic generators need to be defined in addition to normal traffic nodes. At the same time, it is necessary to define to which target nodes each traffic generator will apply what class of traffic, and the size of the traffic applied.
Two traffic generators are defined in the training scenario topology as shown in fig. 1, each traffic generator sending a specified traffic to a particular service node. The traffic generator 1 applies interference traffic to the terminals 1, 2, 3 and 4, the traffic generator 2 applies interference traffic to the terminals 6 and 7, and the network relay node through which the traffic generator 2 passes includes the router 1, 2 and 3.
Assume that traffic generator 1, router 1 and terminal 1, respectively, are successfully virtualized at 3 different physical compute nodes. Then the traffic from the traffic generator 1 to the terminal 1 will first pass through the physical node 1 where the traffic generator 1 is located, and then reach the switch; the physical node 2 where the router 1 is located is reached through the switch; the traffic is then forwarded from the physical node 2 to the switch again; finally the exchange forwards the traffic to the physical node 3 where the terminal 1 is located.
In the above example, the switch carries a traffic size of 4 times the traffic generator. If the defined traffic in the training topology is larger, the switch frequently replicates and forwards the traffic, the upper limit of the performance bottleneck of the switch is easily reached, and meanwhile, the network congestion is easily caused by the large traffic.
From the above, the virtualized nodes in the existing network target resource allocation scheme are not considered in the flow factor when being allocated to each computing node. In the actual test process, the flow is communicated among different physical computing nodes through a switch, so that network storm is easily caused; and the switch frequently duplicates and forwards unnecessary traffic, and the upper limit of the performance bottlenecks of the switch and the network is easy to be reached.
Disclosure of Invention
The invention aims to: aiming at the problems in the prior art, the invention aims to provide a network target range resource allocation method and system based on traffic affinity, so as to reduce traffic cross-physical machine interaction and avoid influencing an actual physical network.
The technical scheme is as follows: in order to achieve the aim of the invention, the invention adopts the following technical scheme:
a network target range resource allocation method based on traffic affinity, comprising:
step 1: analyzing the scene topology of the shooting range, and generating a directed graph based on the flow generator by taking the flow generator as a starting point and taking a target node of the flow generator as an end point;
step 2: calculating the flow size and the virtualization cost of each topological segment of the directed graph;
step 3: sequencing all topology segments according to the order of priority traffic from big to small and then virtualization cost from low to high to obtain a topology segment list A;
step 4: sequencing all the computing nodes according to the sequence of the residual resources from more to less to obtain a computing node list B;
step 5: sequentially taking each element in the computing node list B, traversing the topology segment list A for one of the computing nodes X, taking a topology segment of which the first virtualization cost is in the receiving range of the computing node X for instantiation, removing the topology segment from the topology segment list A, and jumping to the step 6; if the calculation node X does not meet the virtualization cost of any topological segment in the A, adding all nodes in the A into an isolated node list, and jumping to the step 7;
step 6: for a computing node X, recording a set S of virtual nodes related to a topology segment which is currently instantiated on the node, and sequentially searching the topology segments taking the nodes in the set S as starting points in the A; if A is empty, jumping to the step 7; if the topology segment taking the node in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6;
if the topology segments taking the nodes in the set S as the starting points are found, the topology segments are orderly sequenced from high to low according to the virtualization cost to form a list C, and the topology segments with the virtualization cost within the receiving range of the computing node X are orderly searched in the list C; if a topological segment meeting the condition is found, instantiating a node needing to be virtualized in the topological segment, and repeating the step 6; otherwise, adding an isolated node list to the non-instantiated node which takes any node in the set S as a starting point or an ending point and is involved in the non-instantiated topology segment in the list A, removing the corresponding topology segment from the list A, reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6;
step 7: and adding virtual nodes irrelevant to the directed graph in the scene topology into an isolated node list, and instantiating the isolated node list.
Preferably, in the directed graph generated in step 1, each target node records the planned traffic size.
Preferably, the method for calculating the flow size of the topology segment in the step 2 is as follows: if the end point of the topology segment is the network forwarding equipment, the flow of the topology segment is equal to the sum of the flows of all topology segments taking the network forwarding equipment as the starting point; if the end point of the topology segment is the terminal equipment, the flow of the topology segment is equal to the flow defined by the terminal equipment.
Preferably, the virtualization cost of the topology segment in the step 2 is the sum of the CPUs, the sum of the memories and the sum of the disks of the two nodes of the starting point and the ending point of the topology segment.
Preferably, the method for instantiating the orphan node list in step 7 is as follows: and sequentially selecting the isolated nodes from high to low according to the virtualization cost to carry out virtualization, and instantiating the selected isolated nodes into the computing nodes with the largest residual resources.
Preferably, the virtualization cost comprises a CPU, a memory and a disk, the memory is preferentially considered in sorting, the CPU is considered, and the disk is considered finally.
Based on the same inventive concept, the network target range resource distribution system based on traffic affinity provided by the invention comprises:
the topology analysis module is used for analyzing the scene topology of the shooting range, taking the flow generator as a starting point, taking a target node of the flow generator as an end point and generating a directed graph based on the flow generator;
the list generation and sequencing module is used for calculating the flow size and the virtualization cost of each topological segment of the directed graph; sequencing all topology segments according to the order of priority traffic from big to small and then virtualization cost from low to high to obtain a topology segment list A; sequencing all the computing nodes according to the sequence of the residual resources from more to less to obtain a computing node list B;
the directed graph virtualization module is used for executing the virtualization process of the directed graph according to the following steps:
step a: taking each element in the computing node list B in turn, traversing the topology segment list A for one of the computing nodes X, taking the topology segment of which the first virtualization cost is in the receiving range of the computing node X for instantiation, removing the topology segment from the A, and jumping to the step B; if the computing node X does not meet the virtualization cost of any topological segment in the A, adding all nodes in the A into an isolated node list, and ending the directed graph virtualization process;
step b: for a computing node X, recording a set S of virtual nodes related to a topology segment which is currently instantiated on the node, and sequentially searching the topology segments taking the nodes in the set S as starting points in the A; if A is empty, ending the directed graph virtualization process; if the topology segment taking the node in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
if the topology segments taking the nodes in the set S as the starting points are found, the topology segments are orderly sequenced from high to low according to the virtualization cost to form a list C, and the topology segments with the virtualization cost within the receiving range of the computing node X are orderly searched in the list C; if a topological segment meeting the condition is found, instantiating a node needing to be virtualized in the topological segment, and repeating the step b; otherwise, adding an isolated node list to the non-instantiated node which takes any node in the set S as a starting point or an ending point in the list A and is involved in the non-instantiated topology segment, removing the corresponding topology segment from the list A, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
and the isolated node virtualization module is used for adding virtual nodes irrelevant to the directed graph in the scene topology into an isolated node list after the directed graph virtualization process is finished, and instantiating the isolated node list.
Based on the same inventive concept, the invention provides a computer system, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the steps of the network target range resource allocation method based on flow affinity when being loaded to the processor.
The beneficial effects are that: compared with the prior art, the invention has the following advantages:
1. according to the invention, through carrying out flow affinity analysis on the nodes, physical resources required by virtualization are comprehensively considered, and the nodes with highest flow correlation are constrained on the same physical machine for virtualization, so that the interaction of flow across the physical machines in the test process is effectively reduced, and the risk of network storm is reduced.
2. The invention can effectively reduce the demands on the switch and the network in the actual physical networking, and can not cause network performance bottleneck because of frequent forwarding of data packets by the switch.
3. Furthermore, for the isolated node, the invention is started on the computing node with most abundant current physical resources in each round of distribution process, so that the load balance of the total computing cluster can be ensured.
Drawings
Fig. 1 is a diagram of an example of a range training scene topology.
Fig. 2 is a general flow chart of a resource allocation method according to an embodiment of the present invention.
Fig. 3 is a directed graph corresponding to the flow generator 1 in fig. 1.
Fig. 4 is a directed graph corresponding to the flow generator 2 of fig. 1.
Detailed Description
The technical scheme of the invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments.
According to the resource allocation method based on the traffic affinity, through traffic affinity analysis on the nodes in the topology, the nodes with highest traffic correlation are restrained to be started on the same physical machine (computing node), so that the cross-physical node interaction of traffic is reduced, the risk of network storm is reduced, and meanwhile, the pressure of forwarding the traffic by the switch is greatly relieved. And further for the non-traffic affinity nodes, the non-traffic affinity nodes are started on the computing node with the most abundant current physical resources in each round of allocation process, so that the load balance of the total computing clusters is ensured.
As shown in fig. 2, a specific execution flow of a resource allocation method based on traffic affinity according to an embodiment of the present invention is as follows.
Step 1, generating a plurality of directed graphs based on flow generators by taking each flow generator as a starting point and taking a target node of the flow generator as an end point based on scene topology; nodes that do not involve traffic are added to the list of non-traffic affinity nodes. For the training topology shown in fig. 1, 2 directed graphs as shown in fig. 3 and 4 would be generated. In fig. 3 and 4, the letter T represents a traffic generator, R represents a network forwarding device such as a router, a switch, etc., letters a to F represent respective terminal devices, and the numbers in the upper right corner on the terminal devices represent traffic sizes defined to the node in the topology planning phase.
And 2, decomposing the topology segment by segment for each directed graph. The traffic size of each segment topology and the cost required for virtualization are calculated.
Specifically, the cost of each segment of topology virtualization is equal to the sum of the CPU, the memory and the disk of two nodes at the starting point and the ending point of the segment of topology. For the topology segment of R1- > A in FIG. 3, the cost of virtualization is:
Cpu=Cpu
R1
+Cpu
A
Disk=Disk
R1
+Disk
A
Memory=Memory
R1
+Memory
A
wherein the method comprises the steps ofCpu R1 、Cpu A The CPU sizes required for node R1 and a virtualization respectively,Disk R1 、Disk A the disk sizes required for node R1 and a virtualization respectively,Memory R1 、Memory A the memory sizes required for node R1 and a virtualization, respectively.
The flow calculation rule for each topology segment is as follows:
the target node is a network forwarding device, and the flow of the topology segment taking the node as an end point is equal to the sum of the flows of all topology segments taking the node as a starting point;
Traffic(T1:R1)=Traffic(R1:A)+Traffic(R1:R2)+Traffic(R1:B)
the target node is not network forwarding equipment, and the flow of the topology section taking the node as the end point is equal to the flow defined by the node; the traffic from R1 to terminal a is defined as follows:
Traffic(R1:A)=5
the basic unit of the flow size can be Mbps or Gbps, or the customized flow unit can indicate the relative flow size.
And 3, ordering all topology segments from large to small according to the flow, and ordering topology segments with equal flow sizes from low to high according to the requirements of memory, CPU and magnetic disk, wherein the formed topology Duan Lie is denoted as A.
And 4, arranging the residual resources of all the computing nodes in descending order according to the memory and the CPU, and marking the formed computing node list as B. The machine number of the compute node is represented in the form of M1, M2 … MX.
Step 5, sequentially taking each element in the computing node list B, and for one computing node X, executing the following operations:
step 5.1, traversing a topology segment list A, taking a topology segment with a first virtualization cost within the receiving range of a computing node X for instantiation, and removing the topology segment from the list A after the instantiation is completed; step 6, jump.
Step 5.2, if X does not meet the virtualization cost of any topological segment in the topological segment A, adding all nodes in the list A into an isolated node list; step 7, jumping.
Step 6, for the computing node X, executing the following operations:
step 6.1, recording the set of virtual nodes involved in the topology segment that has been currently instantiated on the compute node X. Assuming that the topology segment (T2: F) is first instantiated in step 5.1, the set of virtual nodes involved on the current computing node X is s= (T2, F). Sequentially searching topology segments taking points (T2, F) in the set S as starting points in the list A; if the list A is empty, jumping to the step 7; and if the topology segment taking the point in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6.
And 6.2, if a topological segment taking the point in S as a starting point is found, ordering according to the order from high to low of the virtualization cost to form a list C. Sequentially searching topology segments with virtualization cost within the receiving range of the computing node X in the list C for instantiation, and repeating the steps 6.1 to 6.2 if the topology segments meeting the conditions can be found and the nodes needing virtualization in the instantiated topology segments; otherwise jump to step 6.3.
And 6.3, adding the non-instantiated nodes which are related in the non-instantiated topological segments in the list A and take any element in the set S as a starting point or an ending point into an isolated node list, and removing the topological segments from the list A. And (5) reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6.
And 7, adding non-traffic affinity nodes irrelevant to the traffic directed graph in the scene topology into the isolated node list, and instantiating the isolated node list. The method comprises the following specific steps:
step 7.1, sorting the isolated node list from high to low according to the memory and CPU occupation; and sequentially selecting nodes Y in the isolated node list for virtualization.
And 7.2, reordering all the computing nodes according to the residual resources.
And 7.3, selecting the physical machine with the largest residual resources in the computing node, and executing the virtualization of the node Y. If the virtualization is successful, repeating the steps 7.2 and 7.3, otherwise prompting that the scene is failed to start because of insufficient resources.
For the topologies shown in fig. 3 and 4, the execution is as shown in the following table.
TABLE 1 resource Allocation procedure
Wherein the first column represents the traffic size carried by each topology segment; the numbers of the first row represent the various stages of resource allocation, each stage instantiating a topology segment based on a greedy algorithm system.
Based on the same inventive concept, the network target range resource allocation system based on traffic affinity disclosed in the embodiment of the invention comprises:
the topology analysis module is used for analyzing the scene topology of the shooting range, taking the flow generator as a starting point, taking a target node of the flow generator as an end point and generating a directed graph based on the flow generator;
the list generation and sequencing module is used for calculating the flow size and the virtualization cost of each topological segment of the directed graph; sequencing all topology segments according to the order of priority traffic from big to small and then virtualization cost from low to high to obtain a topology segment list A; sequencing all the computing nodes according to the sequence of the residual resources from more to less to obtain a computing node list B;
the directed graph virtualization module is used for executing the virtualization process of the directed graph according to the following steps:
step a: taking each element in the computing node list B in turn, traversing the topology segment list A for one of the computing nodes X, taking the topology segment of which the first virtualization cost is in the receiving range of the computing node X for instantiation, removing the topology segment from the A, and jumping to the step B; if the computing node X does not meet the virtualization cost of any topological segment in the A, adding all nodes in the A into an isolated node list, and ending the directed graph virtualization process;
step b: for a computing node X, recording a set S of virtual nodes related to a topology segment which is currently instantiated on the node, and sequentially searching the topology segments taking the nodes in the set S as starting points in the A; if A is empty, ending the directed graph virtualization process; if the topology segment taking the node in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
if the topology segments taking the nodes in the set S as the starting points are found, the topology segments are orderly sequenced from high to low according to the virtualization cost to form a list C, and the topology segments with the virtualization cost within the receiving range of the computing node X are orderly searched in the list C; if a topological segment meeting the condition is found, instantiating a node needing to be virtualized in the topological segment, and repeating the step b; otherwise, adding an isolated node list to the non-instantiated node which takes any node in the set S as a starting point or an ending point in the list A and is involved in the non-instantiated topology segment, removing the corresponding topology segment from the list A, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
and the isolated node virtualization module is used for adding virtual nodes irrelevant to the directed graph in the scene topology into an isolated node list after the directed graph virtualization process is finished, and instantiating the isolated node list.
Based on the same inventive concept, the embodiment of the invention discloses a computer system, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the steps of the network range resource allocation method based on traffic affinity when being loaded to the processor.
Claims (9)
1. A method for allocating network target resources based on traffic affinity, comprising:
step 1: analyzing the scene topology of the shooting range, and generating a directed graph based on the flow generator by taking the flow generator as a starting point and taking a target node of the flow generator as an end point;
step 2: calculating the flow size and the virtualization cost of each topological segment of the directed graph; the virtualization cost of the topology segment is the sum of the CPU, the sum of the memory and the sum of the disk of the two nodes of the starting point and the end point of the topology segment;
step 3: sequencing all topology segments according to the order of priority traffic from big to small and then virtualization cost from low to high to obtain a topology segment list A;
step 4: sequencing all the computing nodes according to the sequence of the residual resources from more to less to obtain a computing node list B;
step 5: sequentially taking each element in the computing node list B, traversing the topology segment list A for one of the computing nodes X, taking a topology segment of which the first virtualization cost is in the receiving range of the computing node X for instantiation, removing the topology segment from the topology segment list A, and jumping to the step 6; if the calculation node X does not meet the virtualization cost of any topological segment in the A, adding all nodes in the A into an isolated node list, and jumping to the step 7;
step 6: for a computing node X, recording a set S of virtual nodes related to a topology segment which is currently instantiated on the computing node X, and sequentially searching the topology segments taking the nodes in the set S as starting points in A; if A is empty, jumping to the step 7; if the topology segment taking the node in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6;
if the topology segments taking the nodes in the set S as the starting points are found, the topology segments are orderly sequenced from high to low according to the virtualization cost to form a list C, and the topology segments with the virtualization cost within the receiving range of the computing node X are orderly searched in the list C; if a topological segment meeting the condition is found, instantiating a node needing to be virtualized in the topological segment, and repeating the step 6; otherwise, adding an isolated node list to the non-instantiated node which takes any node in the set S as a starting point or an ending point and is involved in the non-instantiated topology segment in the list A, removing the corresponding topology segment from the list A, reordering all the computing nodes according to the residual resources, and repeating the steps 5 to 6;
step 7: and adding virtual nodes irrelevant to the directed graph in the scene topology into an isolated node list, and instantiating the isolated node list.
2. The traffic affinity based network range resource allocation method according to claim 1, wherein each target node records a planned traffic size in the directed graph generated in step 1.
3. The traffic affinity-based network range resource allocation method according to claim 1, wherein the traffic size calculation method of the topology segment in step 2 is as follows: if the end point of the topology segment is the network forwarding equipment, the flow of the topology segment is equal to the sum of the flows of all topology segments taking the network forwarding equipment as the starting point; if the end point of the topology segment is the terminal equipment, the flow of the topology segment is equal to the flow defined by the terminal equipment.
4. The traffic affinity based network range resource allocation method according to claim 1, wherein the method of instantiating the orphan node list in step 7 is: and sequentially selecting the isolated nodes from high to low according to the virtualization cost to carry out virtualization, and instantiating the selected isolated nodes into the computing nodes with the largest residual resources.
5. The traffic affinity based network range resource allocation method of claim 1, wherein the virtualization cost comprises a CPU, memory and disk, and the memory is prioritized during the ordering, the CPU is considered next, and the disk is considered last.
6. A network target resource allocation system based on traffic affinity, comprising:
the topology analysis module is used for analyzing the scene topology of the shooting range, taking the flow generator as a starting point, taking a target node of the flow generator as an end point and generating a directed graph based on the flow generator;
the list generation and sequencing module is used for calculating the flow size and the virtualization cost of each topological segment of the directed graph; the virtualization cost of the topology segment is the sum of the CPU, the sum of the memory and the sum of the disk of the two nodes of the starting point and the end point of the topology segment; sequencing all topology segments according to the order of priority traffic from big to small and then virtualization cost from low to high to obtain a topology segment list A; sequencing all the computing nodes according to the sequence of the residual resources from more to less to obtain a computing node list B;
the directed graph virtualization module is used for executing the virtualization process of the directed graph according to the following steps:
step a: taking each element in the computing node list B in turn, traversing the topology segment list A for one of the computing nodes X, taking the topology segment of which the first virtualization cost is in the receiving range of the computing node X for instantiation, removing the topology segment from the A, and jumping to the step B; if the computing node X does not meet the virtualization cost of any topological segment in the A, adding all nodes in the A into an isolated node list, and ending the directed graph virtualization process;
step b: for a computing node X, recording a set S of virtual nodes related to a topology segment which is currently instantiated on the computing node X, and sequentially searching the topology segments taking the nodes in the set S as starting points in A; if A is empty, ending the directed graph virtualization process; if the topology segment taking the node in the set S as the starting point cannot be found, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
if the topology segments taking the nodes in the set S as the starting points are found, the topology segments are orderly sequenced from high to low according to the virtualization cost to form a list C, and the topology segments with the virtualization cost within the receiving range of the computing node X are orderly searched in the list C; if a topological segment meeting the condition is found, instantiating a node needing to be virtualized in the topological segment, and repeating the step b; otherwise, adding an isolated node list to the non-instantiated node which takes any node in the set S as a starting point or an ending point in the list A and is involved in the non-instantiated topology segment, removing the corresponding topology segment from the list A, reordering all the computing nodes according to the residual resources, and repeating the steps a to b;
and the isolated node virtualization module is used for adding virtual nodes irrelevant to the directed graph in the scene topology into an isolated node list after the directed graph virtualization process is finished, and instantiating the isolated node list.
7. The traffic affinity based network range resource allocation system according to claim 6, wherein in the list generation and ordering module, the traffic size of the topology segment is calculated by: if the end point of the topology segment is the network forwarding equipment, the flow of the topology segment is equal to the sum of the flows of all topology segments taking the network forwarding equipment as the starting point; if the end point of the topology segment is the terminal equipment, the flow of the topology segment is equal to the flow defined by the terminal equipment.
8. The traffic affinity based network target resource allocation system according to claim 6, wherein the method of instantiating the orphan node list in the orphan node virtualization module is: and sequentially selecting the isolated nodes from high to low according to the virtualization cost to carry out virtualization, and instantiating the selected isolated nodes into the computing nodes with the largest residual resources.
9. A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when loaded to the processor implements the steps of the traffic affinity based network range resource allocation method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311006748.8A CN116723143B (en) | 2023-08-11 | 2023-08-11 | Network target range resource allocation method and system based on traffic affinity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311006748.8A CN116723143B (en) | 2023-08-11 | 2023-08-11 | Network target range resource allocation method and system based on traffic affinity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116723143A CN116723143A (en) | 2023-09-08 |
| CN116723143B true CN116723143B (en) | 2023-11-10 |
Family
ID=87866514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311006748.8A Active CN116723143B (en) | 2023-08-11 | 2023-08-11 | Network target range resource allocation method and system based on traffic affinity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116723143B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117112234B (en) * | 2023-10-19 | 2024-02-09 | 南京赛宁信息技术有限公司 | Reliable resource pre-allocation method and system in network target range |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915407A (en) * | 2016-04-13 | 2016-08-31 | 重庆邮电大学 | Flow generator virtualization realization system and flow generator virtualization realization method based on SDN |
| CN110380906A (en) * | 2019-07-25 | 2019-10-25 | 哈尔滨工业大学 | A kind of mapping method of virtual network of extensive multidimensional fusion |
| CN111600913A (en) * | 2020-07-22 | 2020-08-28 | 南京赛宁信息技术有限公司 | Self-adaptive access method and system for real equipment in attack and defense scene of network shooting range |
| CN113259164A (en) * | 2021-05-18 | 2021-08-13 | 广州锦行网络科技有限公司 | Method for realizing virtual-real networking based on virtual routing system construction |
| CN114024880A (en) * | 2022-01-10 | 2022-02-08 | 南京赛宁信息技术有限公司 | Network target range probe acquisition method and system based on proxy IP and flow table |
| CN114422446A (en) * | 2022-03-29 | 2022-04-29 | 南京赛宁信息技术有限公司 | Application layer background traffic scheduling method and system in target range |
| CN115086250A (en) * | 2022-07-20 | 2022-09-20 | 南京赛宁信息技术有限公司 | Network target range distributed traffic generation system and method |
| CN115208660A (en) * | 2022-07-14 | 2022-10-18 | 软极网络技术(北京)有限公司 | Transparent access method for network target range equipment |
| CN115225508A (en) * | 2022-04-06 | 2022-10-21 | 郑州信大先进技术研究院 | Internet target range scene construction method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765595B (en) * | 2019-10-10 | 2022-12-09 | 内蒙古农业大学 | SDN data center network flow scheduling method based on multilayer virtual topology energy saving |
-
2023
- 2023-08-11 CN CN202311006748.8A patent/CN116723143B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915407A (en) * | 2016-04-13 | 2016-08-31 | 重庆邮电大学 | Flow generator virtualization realization system and flow generator virtualization realization method based on SDN |
| CN110380906A (en) * | 2019-07-25 | 2019-10-25 | 哈尔滨工业大学 | A kind of mapping method of virtual network of extensive multidimensional fusion |
| CN111600913A (en) * | 2020-07-22 | 2020-08-28 | 南京赛宁信息技术有限公司 | Self-adaptive access method and system for real equipment in attack and defense scene of network shooting range |
| CN113259164A (en) * | 2021-05-18 | 2021-08-13 | 广州锦行网络科技有限公司 | Method for realizing virtual-real networking based on virtual routing system construction |
| CN114024880A (en) * | 2022-01-10 | 2022-02-08 | 南京赛宁信息技术有限公司 | Network target range probe acquisition method and system based on proxy IP and flow table |
| CN114422446A (en) * | 2022-03-29 | 2022-04-29 | 南京赛宁信息技术有限公司 | Application layer background traffic scheduling method and system in target range |
| CN115225508A (en) * | 2022-04-06 | 2022-10-21 | 郑州信大先进技术研究院 | Internet target range scene construction method |
| CN115208660A (en) * | 2022-07-14 | 2022-10-18 | 软极网络技术(北京)有限公司 | Transparent access method for network target range equipment |
| CN115086250A (en) * | 2022-07-20 | 2022-09-20 | 南京赛宁信息技术有限公司 | Network target range distributed traffic generation system and method |
Non-Patent Citations (3)
| Title |
|---|
| A Versatile Description Framework for Modeling Behaviors in Traffic Simulations;Alexandre Bonhomme et.al;《2014 IEEE 26th International Conference on Tools with Artificial Intelligence》;全文 * |
| 网络靶场中动态可配置虚拟网络技术研究与实现;安彩虹;《中国优秀硕士学位论文全文数据库（信息科技辑）》(第04期);全文 * |
| 面向未来的网络靶场体系架构研究;刘智国;于增明;王建;张磊;黎康盛;沈辉;;信息技术与网络安全(06);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116723143A (en) | 2023-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108322333B (en) | Virtual network function placement method based on genetic algorithm | |
| CN104993941B (en) | One kind is based on Openflow network high fault tolerance virtual network mapping algorithms | |
| CN116723143B (en) | Network target range resource allocation method and system based on traffic affinity | |
| CN104954270A (en) | Techniques for aggregating hardware routing resources in a multi-packet processor networking system | |
| Xu et al. | A multiple priority queueing genetic algorithm for task scheduling on heterogeneous computing systems | |
| CN112468401B (en) | Network-on-chip routing communication method for brain-like processor and network-on-chip | |
| CN113708972A (en) | Service function chain deployment method and device, electronic equipment and storage medium | |
| CN111371616B (en) | Virtual network function chain deployment method and system for NUMA (non Uniform memory Access) architecture server | |
| Plakunov et al. | Data center resource mapping algorithm based on the ant colony optimization | |
| CN112073237B (en) | Large-scale target network construction method in cloud edge architecture | |
| CN107341054A (en) | Task executing method, device and computer-readable recording medium | |
| Tessier et al. | Topology-aware data aggregation for intensive I/O on large-scale supercomputers | |
| CN110380906B (en) | Large-scale multidimensional fusion virtual network mapping method | |
| CN110191382B (en) | Virtual link priority mapping method based on path sorting | |
| Qing et al. | Hybrid virtual network embedding with K-core decomposition and time-oriented priority | |
| CN113094179B (en) | Job allocation method, job allocation device, electronic equipment and readable storage medium | |
| CN116302453B (en) | Task scheduling method and device for quantum electronic hybrid platform | |
| CN109308210B (en) | Method for optimizing NFV forwarding service chain performance on multi-core server | |
| CN113395183B (en) | Virtual node scheduling method and system for network simulation platform VLAN interconnection | |
| CN114298431A (en) | Network path selection method, device, equipment and storage medium | |
| CN106789289B (en) | Method and apparatus for virtual network mapping | |
| Ezell et al. | I/O router placement and fine-grained routing on Titan to support Spider II | |
| CN112783673A (en) | Method and device for determining call chain, computer equipment and storage medium | |
| Wang et al. | Clustering of virtual network function instances oriented to compatibility in 5G network | |
| JP6339978B2 (en) | Resource allocation management device and resource allocation management method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |