Evaluation method of cluster information dissemination capability in peer-to-peer networks

Abstract

The present invention provides a method for evaluating cluster information dissemination capability in a peer-to-peer network. The peer-to-peer network includes: m nodes and a controller. The cluster composed of each node performs information exchange through the peer-to-peer network. The nodes include: a first-initiating information receiving node, a single-source information receiving node, and a multi-source information receiving node. The controller triggers the first-initiating information receiving node to generate information F w (1) Record the time t1 when the first information receiving node sends the information and the time t2 when each node receives the information 收 The peer-to-peer network stipulates that the single-source information receiving node and the multi-source information receiving node send information F at the delay time dt after receiving the information w (n), dt is the forwarding period, n is the number of forwarding times, and the information propagation efficiency η of the forwarding node is calculated by the formula, #imgabs0#

Description

Technical Field

[0001] The invention belongs to the field of information dissemination algorithms and evaluation algorithms, and in particular relates to an information dissemination capability verification and analysis evaluation algorithm for a cluster in a peer-to-peer network. Background Art

[0002] Information dissemination is one of the important supports for cluster collaborative execution of tasks. Information dissemination methods include limited networks, enumeration networks, and decentralized peer-to-peer networks. Since the decentralized peer-to-peer model is based on network topology, all nodes are peer-to-peer structures, and have the characteristics of simplicity, flexibility, and easy expansion, it is a very important model in cluster information dissemination. In this model, each node traverses the entire network through connections with adjacent peer nodes, and nodes directly interact with each other to achieve information equality, information cohesion, and role equality, which greatly improves the efficiency of resource sharing and information transmission.

[0003] However, the emergent effect of cluster information dissemination brought about by the decentralized peer-to-peer model leads to the lack of cluster information dissemination processing and verification methods, making it difficult to quantify and evaluate, and it is difficult to identify weak points or obstacles in the dissemination network. Summary of the invention

[0004] In order to solve the problem that the information dissemination capability of a cluster is difficult to verify and evaluate in a peer-to-peer network, the invention aims to provide an information dissemination capability verification and analysis evaluation algorithm for a cluster in a peer-to-peer network. By means of an information distribution expansion algorithm for each node in the cluster, a simple encoding of digital signals and decoupling emergence, the verification of the cluster's dissemination capability and the quantitative evaluation of the dissemination efficiency are achieved.

[0005] In order to achieve the invention purpose of this application, this application adopts the following technical solutions:

[0006] The present invention provides a method for evaluating cluster information dissemination capability in a peer-to-peer network. The peer-to-peer network includes: m nodes and a controller, wherein m≥3. Each node forms a cluster. They interact with each other through the peer-to-peer network. The nodes include: a first-initiating information receiving node, a single-source information receiving node, and a multi-source information receiving node. The controller triggers the first-initiating information receiving node to generate information F w (1) And record the time t1 when the first information receiving node sends the information and the receiving time t1 of each node 收 The peer-to-peer network stipulates that the single-source information receiving node and the multi-source information receiving node send the above information F at the delay time dt after receiving the information w (n), dt is the forwarding period, n is the information verification value, that is, the number of forwarding times, if the single-source information receiving node and the multi-source information receiving node do not have the above information F before w (n), then F w (n) is deemed to be the effective amount received, of which:

[0007] (a) The first information receiving node broadcasts information F w (1)

[0008] When there is no information in the peer-to-peer network, the information check value n = 0; the controller triggers a first information receiving node to send a message at time t1, and the first information receiving node information will be F w The information check value of (n) n=0 changes to n=1, and the information F is broadcasted externally. w (1);

[0009] (b) Information processing and forwarding mechanism of single-source information receiving node

[0010] Assume that the single-source information receiving node is at t 收 to 收 +dt period only received information F w (n), at t 收 to 收 +dt period, the single-source information receiving node does not have the above information F before w (n), then the single-source information receiving node will w (n) the check value n becomes 1+n, n is 2, 3, 4, etc.; and at t 收 +dt time to send information F w (n+1) broadcast outward;

[0011] (c) Information processing and forwarding mechanism of multi-source information receiving nodes

[0012] If the multi-source information receiving node is at t 收 to 收 +dt period, the first few F w (n-1), F w (n), F w (n+1)... the same information, compare the size of n-1, n, n+1... check values, then the multi-source information receiving node will add 1 to the minimum check value and then 收 to 收 +dt time, the information Fw(n) is broadcasted outward, n is the minimum check value +1, that is, the above multi-source information receiving node (Z k ) forwarding level, n is 2, 3, 4...;

[0013] (d) Calculate the information transmission efficiency η

[0014] The information propagation efficiency η of single-source information receiving nodes and multi-source information receiving nodes is calculated by formula (1):

[0015]

[0016] Where n is the checksum value broadcasted by the single-source information receiving node and the multi-source information receiving node, that is, the number of forwarding times;

[0017] m is the total number of cluster nodes;

[0018] Δt is the total delay of information propagation, Δt = t 收 +dt-t1,t 收 is the time when the single-source information receiving node or the multi-source information receiving node receives the information, t1 is the time when the first information receiving node sends the information, and dt is the forwarding period;

[0019] The smaller the value of information propagation efficiency η, the stronger its propagation ability.

[0020] The method for evaluating cluster information dissemination capability in a peer-to-peer network of the present invention, wherein: the effective reception amount is information that has not been received before.

[0021] The method for evaluating the cluster information dissemination capability in a peer-to-peer network of the present invention comprises the following steps: in step (b) and step (c), if the above information F w (n), the first information receiving node, the single-source information receiving node and the multi-source information receiving node receive the above information F w (n) Not broadcast.

[0022] The method for evaluating cluster information dissemination capability in a peer-to-peer network of the present invention has the following advantages:

[0023] (1) The application of this method is not limited to the information transmission configuration of cluster nodes, and can be applied to chain and star networks, especially to decentralized peer-to-peer networks;

[0024] (2) The application of this method is not limited by the number of cluster nodes and the level of information distribution transmission.

[0025] (3) The application of this method is not limited to a specific command end and a receiving end. Each node in the cluster can be in duplex mode, and the command end and the receiving end can be any node in the cluster.

[0026] (4) The application of this method is not limited to the direction of information propagation, and information can be sent and received in all directions.

[0027] (5) For information transmission efficiency η, the smaller the value, the higher the transmission efficiency.

[0028] (6) To improve the efficiency of information dissemination, the information forwarding level can be achieved by changing the layout of cluster nodes and reducing the multi-level propagation delay.

[0029] (7) This propagation method achieves emergent decoupling of cluster information propagation in a peer-to-peer network through simple coding operations of digital signals. It can be used as an information propagation distribution algorithm and propagation verification algorithm with good flexibility and universality.

[0030] (8) This information dissemination efficiency method is not limited by the size of the cluster and can achieve quantitative calculation of the dissemination efficiency with a simple method. It can also be used to improve the efficiency of information dissemination.

[0031] (9) The present invention adds an information verification value, which can effectively record the number of forwarding times. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the relationship between nodes and controllers in the method for evaluating cluster information dissemination capability in a peer-to-peer network of the present invention.

[0033] exist Figure 1 In the figure, label 1 is the controller; label 2 is the node; label Z is the w It is the first information receiving node; label Z j is a single-source information receiving node; Z k It is a multi-source information receiving node. DETAILED DESCRIPTION

[0034] For the sake of clarity, the first-source information receiving node, the single-source information receiving node and the multi-source information receiving node are first defined.

[0035] The first information receiving node refers to the source node of the information or the node that receives a certain information for the first time in the network.

[0036] A single-source information node refers to a node that receives only one piece of information within a certain period of time.

[0037] A multi-source information node refers to a node that receives multiple pieces of information within a certain period of time.

[0038] The first information receiving node, the single-source information node and the multi-source information node are random. They are not designated nodes. There is only one first information receiving node in the cluster, and there can be one or more single-source information nodes and multi-source information nodes.

[0039] like Figure 1 As shown, the peer-to-peer network of the present invention includes: m nodes 2 and a controller 1, where m≥3, each node 2 forms a cluster, and they exchange information through the peer-to-peer network. For the convenience of the following description, the nodes are artificially divided into three categories, that is, the nodes 2 include: the first information receiving node Z w , single source information receiving node Z j and multi-source information receiving node Z k, controller 1 triggers the first information receiving node Z w Generate information F w (1) And record the time t1 when the first information receiving node sends the information and the receiving time t1 of each node 收 , the peer-to-peer network agrees on a single-source information receiving node Z j and multi-source information receiving node Z k The above information F is sent at the delay time dt after receiving the information w (n), dt is the forwarding period, n is the information check value, that is, the number of forwarding times. If the single-source information receiving node Z j and multi-source information receiving node Z k The above information did not exist before. w (n), then F w (n) is regarded as the effective reception amount. The method for evaluating the cluster information dissemination capability in a peer-to-peer network of the present invention comprises the following steps:

[0040] (a) Node Z receiving the first information w Broadcast information to the outside world w (1)

[0041] When there is no information in the peer-to-peer network, the information check value n = 0; the controller 1 triggers a first information receiving node Z at time t1. w Send a message, the first message receiving node Z w Information will be F w The information check value of (n) n=0 changes to n=1, and the information F is broadcasted externally. w (1);

[0042] (b) Single-source information receiving node Z j Information processing and forwarding mechanism

[0043] Assume that the single source information receiving node Z j In t 收 to 收 +dt period only received information F w (n), at t 收 to 收 +dt period, single source information receiving node Z j The above information did not exist before. w (n), then the single-source information receiving node Z j F w (n) the check value n becomes 1+n, n is 2, 3, 4, etc.; and at t 收 +dt time to send information F w (n+1) broadcast outward;

[0044] (c) Multi-source information receiving node Z kInformation processing and forwarding mechanism

[0045] If the multi-source information receiving node Z k In t 收 to 收 +dt period, the first few F w (n-1), F w (n), F w (n+1)... the same information, compare the size of n-1, n, n+1... check values, then the multi-source information receiving node Z k After adding 1 to the minimum check value, 收 to 收 +dt time, the information Fw(n) is broadcasted outward, n is the minimum check value +1, that is, the multi-source information receiving node Z k The forwarding level, n is 2, 3, 4...;

[0046] (d) Calculate the information transmission efficiency η

[0047] Single source information receiving node Z j and multi-source information receiving node Z k The information transmission efficiency η is calculated by formula (1):

[0048]

[0049] Where n is the single source information receiving node Z j and multi-source information receiving node Z k The broadcast checksum value, i.e. the number of forwarding times;

[0050] m is the total number of cluster nodes;

[0051] Δt is the total delay of information propagation, Δt = t 收 +dt-t1,t 收 Single source information receiving node Z j or multi-source information receiving node Z k The time of receiving information, t1 is the time when the first information is received by the node Z w The time to send information, dt is the forwarding period;

[0052] The smaller the value of information propagation efficiency η of node 2, the stronger its propagation ability.

[0053] In step (b) and step (c), if the above information F already exists w (n), the first information receiving node Z w , single source information receiving node Z j and multi-source information receiving node Z k For the above information w(n) No broadcast. When a multi-source information receiving node Z k After forwarding the information, the first information receiving node Z w , single source information receiving node Z j and other multi-source information receiving nodes Z k will receive the forwarded information again, but because the first information receiving node Z w , single source information receiving node Z j and multi-source information receiving node Z k The above information already exists w (n), the first information receiving node Z w , single source information receiving node Z j and multi-source information receiving node Z k For the above information w (n) No broadcasting, and no longer counting the above-mentioned first information receiving node Z w , single source information receiving node Z j and multi-source information receiving node Z k The information dissemination efficiency η.

[0054] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, technicians familiar with the field may make various corresponding changes and modifications based on the present invention, but these changes and modifications should all fall within the scope of protection of the claims of the present invention.

Claims

1. A method for evaluating cluster information dissemination capability in a peer-to-peer network, the peer-to-peer network comprising: m nodes (2) and a controller (1), where m≥3, each node (2) forms a cluster, and they exchange information through a peer-to-peer network. The nodes (2) include: the first information receiving node (Z w ), single source information receiving node (Z j ) and multi-source information receiving node (Z k ), the controller (1) triggers the first information receiving node (Z w ) Generate information F w (1) And record the time t1 when the first information receiving node sends the information and the receiving time t1 of each node 收 , peer-to-peer network agrees on a single source information receiving node (Z j ) and multi-source information receiving node (Z k ) sends the above information F at the delay time dt after receiving the information w (n), dt is the forwarding period, n is the information check value, that is, the number of forwarding times, if the single-source information receiving node (Z j ) and multi-source information receiving node (Z k ) The above information did not exist before. w (n), then F w (n) is considered as effective reception, characterized by: (a) The receiving node of the first information (Z w ) Broadcast information to the outside world w (1) When there is no information in the peer-to-peer network, the information check value n=0; the controller (1) triggers a first information receiving node (Z) at time t1. w ) sends a message, the first message receiving node (Z w ) Information will be F w The information check value of (n) n=0 changes to n=1, and the information F is broadcasted externally. w (1); (b), single source information receiving node (Z j ) Information processing and forwarding mechanism Assume that a single-source information receiving node (Z j ) at t 收 to 收 +dt period only received information F w (n), at t 收 to 收 +dt period, the single-source information receiving node (Z j ) The above information did not exist before. w (n), then the single-source information receiving node (Z j ) w (n) the check value n becomes 1+n, n is 2, 3, 4, etc.; and at t 收 +dt time to send information F w (n+1) broadcast outward; (c) Multi-source information receiving node (Z k ) Information processing and forwarding mechanism If the multi-source information receiving node (Z k ) at t 收 to 收 +dt period, the first few F w (q), F w (s), F w (l) ... the same information, compare the size of q, s, l ... check values, then the multi-source information receiving node (Z k ) After adding 1 to the minimum check value, at t 收 to 收 +dt time to send information F w (n) broadcast outward, n is the minimum check value + 1, that is, the multi-source information receiving node (Z k ) forwarding level, n is 2, 3, 4...; (d) Calculate the information transmission efficiency η Single source information receiving node (Z j ) and multi-source information receiving node (Z k The information transmission efficiency η of ) is calculated by formula (1): Where n is the single source information receiving node (Z j ) and multi-source information receiving node (Z k ) The checksum value broadcasted, i.e. the number of forwarding times; m is the total number of cluster nodes; Δt is the total delay of information propagation, Δt = t 收 +dt-t1,t 收 For a single source information receiving node (Z j ) or multi-source information receiving node (Z k ) receives the information, t1 is the time when the first information receiving node (Z w ) is the time to send information, dt is the forwarding period; The smaller the value of information transmission efficiency η of a node (2), the stronger its transmission ability.

2. The method for evaluating cluster information dissemination capability in a peer-to-peer network as claimed in claim 1, characterized in that: In step (b) and step (c), if the above information F already exists w (n), the first information receiving node (Z w ), single source information receiving node (Z j ) and multi-source information receiving node (Z k ) for the above information F w (n) Not broadcast.

Images