CN102694799A - P2P (peer-to-peer) stream media system simulation platform and performance evaluation method - Google Patents
P2P (peer-to-peer) stream media system simulation platform and performance evaluation method Download PDFInfo
- Publication number
- CN102694799A CN102694799A CN2012101573590A CN201210157359A CN102694799A CN 102694799 A CN102694799 A CN 102694799A CN 2012101573590 A CN2012101573590 A CN 2012101573590A CN 201210157359 A CN201210157359 A CN 201210157359A CN 102694799 A CN102694799 A CN 102694799A
- Authority
- CN
- China
- Prior art keywords
- streaming media
- user
- media system
- model
- node
- 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.)
- Granted
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 100
- 238000011156 evaluation Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000008569 process Effects 0.000 claims abstract description 26
- 230000003068 static effect Effects 0.000 claims abstract description 25
- 238000004891 communication Methods 0.000 claims description 24
- 239000012634 fragment Substances 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 230000006399 behavior Effects 0.000 claims description 6
- 230000002452 interceptive effect Effects 0.000 claims description 6
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 6
- 230000008859 change Effects 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a P2P (peer-to-peer) stream media system simulation platform and a performance evaluation method and belongs to the field of simulation of a stream media system built on the basis of a P2P technology. A simulation model disclosed by the invention comprises a static model and a dynamic model; the static model is used for simulating a basic frame of the P2P stream media system; the dynamic model is used for simulating an operation process of the P2P stream media system, and the performance evaluation method is used for evaluating the performance of the P2P stream media system and providing references for deployment of the P2P stream media system. The P2P stream media system is modeled according to the simulation model disclosed by the invention; then the simulation model is operated by a simulation tool to obtain the data related to the performance index, and the result data are analyzed to obtain the change of related performance index for deployment of the P2P stream media system.
Description
Technical Field
The invention belongs to the field of simulation of a streaming media system constructed based on a P2P technology, and particularly relates to a P2P streaming media system simulation platform and a performance evaluation method.
Background
With the popularization of the Internet and the increasing maturity of various network-based applications, the continuous upgrade of the network itself and the software and hardware of users, and the continuous development of multimedia technologies, the demands of people on network resources are no longer limited to characters and images, and multimedia resources such as audio and video become important contents in the network, and the streaming media technology can well support the applications. The streaming media refers to a continuous time-base media format that is transmitted and played in time sequence on a data network such as the Internet or Intranet by adopting a streaming transmission mode. The streaming media system mainly includes three parts, as shown in fig. 1, which are a streaming media server system, a streaming media transmission network system, and a streaming media client system.
Because the streaming media data has the characteristics of real-time property, large data volume, sequentiality and continuity, it is very difficult work to realize large-scale streaming media service under the current network conditions (such as the Internet and a wireless communication network). The existing distributed service simulation platform can only simulate according to a specific service of the internet, but can not simulate the multi-service fusion, and meanwhile, the platforms do not support the simulation of a specific QoS index in a telecommunication network; in addition, the existing telecommunication service network simulation platform cannot simulate the algorithm of the distributed service and the new service.
Disclosure of Invention
The present invention aims to solve the above-mentioned problems in the prior art, and provides a simulation platform and a performance evaluation method for a P2P streaming media system, where the simulation platform is used to model the overall architecture, solution and scheduling algorithm of the P2P streaming media system, and the performance evaluation method is used to evaluate the performance of the P2P streaming media system, so as to provide a reference basis for the deployment of the P2P streaming media system. Modeling is carried out according to the simulation platform, the simulation platform is operated to obtain performance index related data, and the result data is analyzed, so that the change condition of the related performance index deployed by the P2P streaming media system is obtained. The invention can realize the simulation of the multi-service multi-strategy scene on the same platform, and in addition, if the multi-service multi-strategy needs to be expanded in the future, the simulation of the new strategy of the new service can be realized only by singly realizing the new service and the new strategy and adding the new service and the new strategy into the input configuration item of the platform.
The invention is realized by the following technical scheme:
a P2P streaming media system simulation platform, comprising a static model and a dynamic model;
the static model is used for simulating a basic framework of a P2P streaming media system, wherein the basic framework of the P2P streaming media system comprises user nodes, core network element nodes, router nodes and communication links among the three nodes;
the static model comprises four sub-models which are respectively a user model, a node model, a topology model and a link model, and the four sub-models respectively simulate the three types of nodes in the basic framework of the P2P streaming media system and the attributes of communication links among the three types of nodes; wherein,
the user model is used for simulating the behaviors of the user, such as joining and exiting, channel selection and the like;
the node model is used for simulating the conditions of each node in the P2P streaming media system;
the topological model is used for simulating the hierarchical distribution condition of each node in the P2P streaming media system;
the link model is used for simulating the attribute of a communication link between nodes in the P2P streaming media system;
the dynamic model is used for simulating the operation process of the P2P streaming media system; the dynamic model comprises three submodels, namely a control model, a protocol model and a playing model;
the control model is used for simulating the operation logic of the P2P streaming media system;
the protocol model is used for simulating an interactive framework of the P2P streaming media system, namely, communication protocol messages in the running process are defined;
the play model is used for simulating a streaming media play mode reflected for a user in the running process of the P2P streaming media system.
The user model includes:
user arrival model: respectively configuring the number of peak request users and peak time points according to the piecewise stable Poisson distribution, then calculating to obtain the user request quantity in 24 hours according to the number of peak request users and the peak time points by using a Poisson distribution probability function, and inputting the obtained user request quantity in 24 hours into a database of a simulation tool;
user session duration model: according to the lognormal distribution, different session durations are configured for users when the users are online, namely two parameters of the average online time length of the users and the standard deviation of the session time length are respectively configured, and one user session time length can be generated for each online user by bringing the two parameters into the lognormal distribution function; the standard deviation of the session duration is a lognormal distribution scale parameter; the two parameters of the user average online duration and the standard deviation of the session duration are two parameters required in the lognormal distribution probability function,
media resource access model: after the user is online, the total number of accessible channels in the P2P streaming media system is configured for the user according to the Zipff distribution in combination with the number of published channels in the P2P streaming media system on the current day.
The conditions of each node in the P2P streaming media system include:
computing resource data of each node, including the total amount and the allowance of CPU resources;
the communication resource data of each node comprises the total amount and the allowance of communication bandwidth resources and the total amount and the allowance of connection number resources.
The hierarchical distribution of each node in the P2P streaming media system includes:
dividing roles of each node: the roles comprise core network element nodes, router nodes and user nodes, wherein the core network element nodes refer to all servers deployed on the core network side, the router nodes comprise core network routers and access network routers, and the user nodes refer to user terminals;
connection topology of each node: the access network router topology can employ a random topology or a ring topology or a mesh topology or a fully connected topology.
The attributes of the communication link include link delay data and link bandwidth data.
The operation logic refers to a business process and a resource scheduling strategy; the business process comprises a media publishing process and a user access process;
the resource scheduling policy defines data scheduling relationships between user nodes and between a user node and a core network element node, and includes:
and (3) updating the strategy of the neighbor node when the user node joins and exits: when the number of the neighbor nodes of the user node is less than the threshold value, or the service capability of the neighbor nodes is less than the threshold value, or the time from the last time of updating the neighbor nodes exceeds the specified time, the user node updates the neighbor nodes; the threshold value, the specified time and the threshold value are fixed values in the P2P streaming media system or are set by a system user; the neighbor node is: after a user node is online, in order to acquire data fragments from other user nodes watching the same program, acquiring a member list from a Tracker, establishing a neighbor relation with part of user nodes on the member list, wherein the user node successfully establishing the neighbor relation with the user node is a neighbor node of the user node, and two user nodes which are neighbors mutually not only acquire media data fragments from the other side but also provide downloading of local media data fragments for the other side; the Tracker is a resource index server and is a core network element node in a P2P streaming media system;
data scheduling policy on user node in the playing process: adopting a data scheduling algorithm combining pure pull or push pull;
strategy for increasing the number of channel copies as the amount of access increases: selecting a core network element node which is idle at present, and publishing a channel copy to the core network element node;
policy to reduce the number of channel copies when the amount of access is reduced: selecting a core network element node which is busy at present, and deleting a part of channel copies on the core network element node;
the protocol message comprises definition of each field of the protocol, size of a protocol message packet and an interactive flow of the protocol message;
the streaming media playing mode comprises the following steps:
when the required data fragment arrives, starting normal playing;
and if the required data fragment does not arrive, the playing is suspended.
A method for utilizing the simulation platform of the P2P streaming media system to evaluate the performance of the P2P streaming media system comprises the steps of firstly writing all configurations in a static model as input data into a database and a configuration file of a simulation tool, then reading the input data in the database and the configuration file by a program of the P2P streaming media system simulation platform through a dynamic model to carry out simulation operation and result data output, and finally utilizing the result data to evaluate the performance of the P2P streaming media system simulation platform; the program of the P2P streaming media system simulation platform is written by using a simulation tool;
the method comprises the following steps:
(1) instantiating the static model and the dynamic model, and determining a basic framework and an operation flow of the P2P streaming media system; the instantiation refers to inputting and recording the specific configuration of the three types of nodes into the simulation tool, wherein the 24-hour user request amount in the static model is recorded into a database of the simulation tool, and all the other configurations in the static model and all the configurations of the dynamic model are recorded into a configuration file of the simulation tool; the specific configuration refers to the relevant contents of node models, network resources and the like input or selected by a user through a foreground interface;
(2) the program of the P2P streaming media system simulation platform reads the input data in the configuration file and the database, and the control model controls the operation of the program of the P2P streaming media system simulation platform to realize the simulation operation of the P2P streaming media system simulation platform; recording simulation result data in a simulation operation process, wherein the simulation result data comprise timestamps and real-time bandwidth loads on a core network element node and a user node; the data of specific configuration input by the user can be written into a configuration file or a database, so that the data can be called and read by a program of the simulation platform;
(3) calculating to obtain a user experience side evaluation index and a network performance side evaluation index by using the simulation result data obtained in the step (2), wherein the user experience side evaluation index comprises the starting delay and the playing continuity of a user watching channel, and the network performance side evaluation index comprises the bandwidth utilization rate and the neighbor contribution rate;
the starting time delay refers to a time interval from the time when the user node sends a program playing request to the time when the user node starts playing the program, namely, the difference between a timestamp for starting playing of the user node and a timestamp for online of the user node; the lower the index value is, the faster the playing is started, and the shorter the user waits for playing time;
the playing continuity refers to that after the streaming media starts playing, in a time period T, the time for which the streaming media can be continuously played on the user node is T ', and the playing continuity is T'/T, namely the percentage of the time length of continuous playing to the playing speed of the user; wherein, T is the difference of two timestamps from the beginning of playing to the current time, T' is the sum of all the time segments of continuous playing in the time segment T, and each time segment of continuous playing refers to the difference of two timestamps from playing to pausing; the higher this property the more continuous the viewing;
the bandwidth utilization rate refers to the percentage of the bandwidth used by the core network element node in the available bandwidth and the percentage of the bandwidth used by the user node in the available bandwidth in a sampling time period, that is, the real-time bandwidth load on the core network element node and the user node accounts for the percentage of the configured bandwidth capacity;
the neighbor contribution rate is: the user node provides the proportion of the uploaded neighbor node number to the total neighbor node number, and represents the contribution degree of the user node to the neighbor node;
(4) and completing the performance evaluation of the P2P streaming media system according to the starting time delay, the playing continuity, the bandwidth utilization rate and the neighbor contribution rate.
The step (1) specifically comprises the following steps:
(11) instantiating a static model, comprising:
(A1) determining the computing resource data of each node and the communication resource data of each node of the P2P streaming media system, and instantiating a node model;
(A2) determining the number and connection topology conditions of various nodes of the P2P streaming media system, instantiating a topology model, constructing an overall architecture of the P2P streaming media system by combining the node model, and connecting the nodes in the entire P2P streaming media system through the topology architecture according to the configured number and connection topology of the nodes;
(A3) determining link delay data and link bandwidth data of the P2P streaming media system, instantiating a link model, and loading link attributes through configuration about link information in a configuration file in combination with the constructed overall architecture;
(A4) instantiating a user model according to a user arrival model, a user session duration model and a media resource access model; the three models are provided for the user configuration of the P2P streaming media system, the user configures the required parameters according to the requirements, and the P2P streaming media system simulation platform can provide the simulation of the corresponding user behaviors;
(12) instantiating a dynamic model, comprising:
(B1) instantiating a control model according to the operational logic of the P2P streaming media system;
(B2) instantiating a protocol model according to protocol messages of the P2P streaming media system;
(B3) the playback model is instantiated according to the streaming media playback mode of the P2P streaming media system.
The simulation tool adopts OMNeT + + or OPNeT.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention is oriented to a novel mobile service network, can simulate the multi-service fusion, supports the simulation of specific QoS indexes in a telecommunication network, and can simulate the algorithm of distributed services and new services.
(2) In the static model, the simulation of multiple strategies can be realized through the configuration of the strategies. The dynamic model can realize multi-service simulation by configuring the service flow;
(3) if a multi-service multi-strategy needs to be expanded in the future, the new service and the new strategy only need to be independently coded and realized, the new service and the new strategy are added into an input configuration item of the strategy in a simulation platform, and the simulation of the new strategy of the new service can be realized by selecting the new configuration.
Drawings
Fig. 1 is a schematic structural diagram of a P2P streaming media system simulation platform according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
in order to provide large-scale streaming media services on the existing network, the present invention analyzes and researches various aspects of the streaming media system, including the system architecture, the media data encoding technology, the proxy cache and multicast technology, etc., and focuses on the architecture and the operation mechanism of the streaming media system.
Specifically, a P2P streaming media system simulation platform, as shown in fig. 1, includes two major parts, which are a static model and a dynamic model. The program of the simulation platform of the invention provides a web page for the user, and the user can configure each model through the configuration options in the web page. The 24-hour user request volume in the static model is put into the database, and other configurations in the static model and configurations of all dynamic models (such as total user node volume, user node, server and router bandwidth, topology among routers, service policy configuration, etc.) are all input into the configuration file.
1, static model
The static model further comprises four submodels, respectively: user models, node models, topology models, and link models.
(1) The user model comprises the following steps: for simulating the behaviour of a user, i.e. for simulating a specific user situation, the following models are included:
● user arrival model
Respectively configuring the number of peak requesting users and the peak time point according to the piecewise stable Poisson distribution (one type of Poisson distribution), then calculating the user request volume in 24 hours according to the number of peak requesting users and the peak time point by using a Poisson distribution probability function, and writing the user request volume in 24 hours into a database of a simulation tool to complete the configuration.
● user session duration model
According to the lognormal distribution, when a user is online, different session durations are configured for the user, namely, the average online duration of the user and a lognormal distribution scale parameter (namely, a standard deviation of session duration) are respectively configured, and the two parameters are two parameters required in a lognormal distribution probability function and therefore need to be configured. The two parameters are brought into the lognormal distribution function to generate an online time length, namely a user conversation time length, for each online user.
● media resource access model
According to the zipov distribution, after the user is online, the total number of channels in the system (which means the number of channels that can be accessed by the user) is configured for the user according to the number of channels already published in the P2P streaming media system on the day.
(2) And (3) node model: the method is used for simulating the conditions of each node in a P2P streaming media system and comprises the following contents:
●, computing resource data for each node, comprising:
CPU resource total and margin;
● communication resource data of each node, comprising:
total amount of communication bandwidth resources and margin;
join the total amount of resources and the margin. The number of connections is the number of currently established connections, and is used as a factor for estimating the current service capability of a node, for example, a node may establish up to 10 connections, and if 5 connections are currently established, the current resource usage amount may be considered to be 50%.
(3) Topological model: the method is used for simulating the hierarchical distribution of the nodes in the P2P streaming media system and comprises the following contents:
● dividing the roles of each node, including core network element node, core network router, access network router and user node;
● connection topology of each node
Access network router topology
● random topology
● Ring topology
● mesh topology
● full connection topology
At present, only the access network routing topology realizes various topology schemes, users can select the topology, the topologies of other nodes are fixed in a program, and the users cannot change the topology.
(4) And (3) link model: the attributes for simulating the communication link between each node in the P2P streaming media system include the following:
● link delay data;
● link bandwidth data.
2, dynamic model
The dynamic model further comprises three submodels, respectively: a control model, a protocol model, and a play model.
(1) A control model: simulating the operating logic of the P2P streaming media system, comprising the following:
● Business Process: the corresponding business process can be automatically operated by selecting the business to be operated;
media publishing process
User visit Process
● resource scheduling policy: defining data scheduling relations among user nodes and between the user nodes and core network element nodes, and designing the following strategies:
when the user node joins and exits, the neighbor node updates the strategy: when the number of the user neighbor nodes is less than a certain threshold value, the service capability of the neighbor nodes is less than a certain threshold value, or the time from the last time of updating the neighbor nodes exceeds a certain specified time, the user nodes update the neighbor nodes. The threshold, the predetermined time and the threshold are currently fixed values in the P2P streaming media system, and may be designed to be set later by a user of the P2P streaming media system, and these values are reasonable values determined according to some research on the current network. The neighbor node is: after a user node is online, in order to acquire data fragments from other user nodes watching the same program, acquiring a member list from a Tracker (the Tracker is a resource index server and belongs to a core network element node in a P2P streaming media system), establishing a neighbor relation with part of user nodes on the member list, wherein the user nodes successfully establishing the neighbor relation with the user nodes are neighbor nodes of the user nodes, and two user nodes which are neighbors mutually acquire media data fragments from opposite ends and provide downloading of local media data fragments for opposite sides;
in the playing process, a node data scheduling strategy: the method comprises a combination of pure pull and push pull data scheduling algorithms (please refer to the clean 'node capability-based P2P streaming media push-pull combined data scheduling algorithm', Chinese scientific paper on-line. 2011.11.16).
Strategy to increase the number of channel copies as the amount of visits increases: and selecting a node which is idle at present, and publishing the channel copy to the node.
Strategy for reducing the number of channel copies as the amount of visits decreases: selecting a node which is busy currently, and deleting some channels on the node (some are judged by a designed algorithm, specifically, the calculated result of the algorithm may be different according to the current environment). Busy and idle are defined by setting some thresholds, such as nodes that are considered busy with bandwidth utilization above 70%.
(2) Protocol model: the interactive framework of the simulation P2P streaming media system, namely the custom protocol message, comprises the following contents:
● protocol field definitions;
● protocol message packet size;
● interaction flow (i.e., protocol) of protocol messages.
These are well defined in the program.
(3) And (3) playing the model: the streaming media playing mode of the simulated P2P streaming media system comprises the following contents:
● when the required data fragment arrives, start playing normally;
● the desired data slice does not arrive and playback is paused.
These are well defined in the program.
A method for evaluating the performance of a P2P streaming media system by using the simulation model comprises the following steps:
the simulation operation of the P2P streaming media system based on the simulation model will generate a series of simulation result information (such as real-time bandwidth load data, playing time stamp data, etc. of each node). The invention provides two types of evaluation indexes based on the simulation result information, namely a user experience side evaluation index and a network performance side evaluation index, and the two types of evaluation indexes are as follows:
1, evaluating indexes at user experience side
A, starting time delay: and sending a program playing request from the user node to the time interval when the user node starts playing the program. The lower the index value is, the faster the playback is started, and the shorter the user waits for the playback time.
B, playing continuity: after the streaming media starts playing, in the time period T, the time for which the streaming media can be continuously played on the user node is T ', and then the playing continuity is defined as T'/T. Wherein, T is the difference between two timestamps from the beginning of playing to the current time, and T' is the total time of continuously playing the streaming media from the beginning of playing to the current time, and is the sum of all the continuously playing timestamps; the playing continuity of each node represents the playing situation of each node, and may be different. The higher this property the more continuous the viewing.
2, network performance side evaluation index
A, bandwidth utilization rate: the bandwidth used by the core network element node as a percentage of the available bandwidth and the bandwidth used by the user node as a percentage of the available bandwidth during a particular time period (e.g., a sample time period).
B, neighbor contribution rate: after the streaming media user node is online, acquiring a neighbor list from a Tracker in order to acquire data fragments from other user nodes watching the same program, and establishing a neighbor relation with other user nodes; two user nodes which are adjacent to each other not only obtain the media data fragments from the opposite end, but also provide the downloading of the local media data fragments for the opposite end; therefore, the contribution rate of the user node, that is, the contribution rate of the user node to the neighbor node, is expressed by using the neighbor contribution rate, that is, the proportion of the number of the neighbor nodes that the user node provides the upload to the total neighbor nodes.
The performance evaluation method comprises the following steps:
step 1: determining node computing resources (CPU) and communication resources (communication bandwidth and connection number) of the P2P streaming media system to configure the capability of the node, support the node to complete corresponding functions and instantiate a node model;
determining the number of various nodes and connection topology of the P2P streaming media system, instantiating a topology model, constructing the whole framework of the P2P streaming media system by combining the node model, and connecting the nodes in the whole network through a certain topology framework according to the configured number of the nodes and the connection topology;
determining the communication link delay and bandwidth of the P2P streaming media system, instantiating a link model, and loading link attributes through configuration on link information in a configuration file in combination with the constructed overall architecture;
and instantiating the user model according to the user arrival model, the user session duration model and the media resource access model. The three models are provided for the system user to configure, the user configures the required parameters according to the requirements, and the simulation model can provide the simulation of the corresponding user behavior.
Step 2, instantiating a control model according to the operation logic of the P2P streaming media system, namely according to four strategies in a service flow and a resource scheduling algorithm, wherein the instantiation comprises a threshold value, a threshold value and a specific flow of the strategies;
according to the communication protocol of the P2P streaming media system, including the definition of each field of the protocol, the size of the protocol message packet, the interactive flow of the protocol message, instantiating a protocol model;
according to the play model of the P2P streaming media system, the required data fragment is reached and the data is played normally; the required data fragment does not arrive, the playing is suspended, and a playing model is instantiated, specifically, the required data fragment arrives and is normally played; the playing process of the P2P streaming media system is controlled by the principle of playing pause when the required data segment does not arrive, thereby forming a playing model of the P2P streaming media system, and the instantiated written data is the relevant timestamp for the user to acquire the data segment.
The steps complete the instantiation of the simulation model, and determine the basic framework and the operation flow of the P2P streaming media system.
Step 3, using a simulation tool (such as OMNeT + +, OPNeT and the like) to complete the configuration of simulation input according to the static model, using a background simulation program (namely a program of a P2P streaming media system simulation platform written on the simulation tool) to control the running of the background program (namely the background program is realized according to a business process and a resource scheduling algorithm in a control model, the control model determines the running of the background program) by reading input data in a configuration file and a database, during the simulation operation, simulation result data (such as registration of nodes, relevant playing time stamps, real-time bandwidth load, calculation load and the like) are recorded, the recording is realized by acquiring the simulation result data through a written program and writing the simulation result data into a database, these result data are extracted as required by the specific formula for calculating the index defined in the present invention. The models in the invention are realized by programming on a simulation tool, and the simulation tool is only used tools and provides an environment for programming and running the whole set of programs. The simulation tool realizes the simulation of partial models, and the simulation of other models is realized by utilizing the simulation platform program written on the simulation tool.
And 4, calculating to obtain the starting delay and the playing continuity of the channel watched by the user and the node bandwidth utilization rate based on the simulation result.
And step 5, finishing the performance evaluation of the P2P streaming media system according to the starting time delay, the playing continuity, the bandwidth utilization rate and the neighbor contribution rate. The performance evaluation is mainly realized by comparing performance indexes under different scenes, the absolute value of the performance indexes has little significance, and the relative values of the performance indexes are focused.
The above-described embodiment is only one embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and variations can be easily made based on the application and principle of the present invention disclosed in the present application, and the present invention is not limited to the method described in the above-described embodiment of the present invention, so that the above-described embodiment is only preferred, and not restrictive.
Claims (10)
1. A P2P streaming media system simulation platform is characterized in that: the P2P streaming media system simulation platform comprises a static model and a dynamic model;
the static model is used for simulating a basic framework of a P2P streaming media system, wherein the basic framework of the P2P streaming media system comprises user nodes, core network element nodes, router nodes and communication links among the three nodes;
the static model comprises four sub-models which are respectively a user model, a node model, a topology model and a link model, and the four sub-models respectively simulate the three types of nodes in the basic framework of the P2P streaming media system and the attributes of communication links among the three types of nodes; wherein,
the user model is used for simulating the behavior of a user;
the node model is used for simulating the conditions of each node in the P2P streaming media system;
the topological model is used for simulating the hierarchical distribution condition of each node in the P2P streaming media system;
the link model is used for simulating the attribute of a communication link between nodes in the P2P streaming media system;
the dynamic model is used for simulating the operation process of the P2P streaming media system; the dynamic model comprises three submodels, namely a control model, a protocol model and a playing model;
the control model is used for simulating the operation logic of the P2P streaming media system;
the protocol model is used for simulating an interactive framework of the P2P streaming media system, namely, communication protocol messages in the running process are defined;
the play model is used for simulating a streaming media play mode reflected for a user in the running process of the P2P streaming media system.
2. The P2P streaming media system emulation platform of claim 1, wherein: the user model includes:
user arrival model: respectively configuring the number of peak request users and peak time points according to the piecewise stable Poisson distribution, and then calculating according to the number of the peak request users and the peak time points by using a Poisson distribution probability function to obtain a user request amount of 24 hours;
user session duration model: when a user is online, two parameters, namely average online time length of the user and standard deviation of session time length, are respectively configured, and the two parameters are brought into a lognormal distribution function to generate a user session time length for each online user;
media resource access model: after the user is online, the total number of accessible channels in the P2P streaming media system is configured for the user according to the Zipff distribution in combination with the number of published channels in the P2P streaming media system on the current day.
3. The P2P streaming media system emulation platform of claim 2, wherein: the conditions of each node in the P2P streaming media system include:
computing resource data of each node, including the total amount and the allowance of CPU resources;
the communication resource data of each node comprises the total amount and the allowance of communication bandwidth resources and the total amount and the allowance of connection number resources.
4. The P2P streaming media system emulation platform of claim 3, wherein: the hierarchical distribution of each node in the P2P streaming media system includes:
dividing roles of each node: the roles comprise core network element nodes, router nodes and user nodes, wherein the core network element nodes refer to all servers deployed on the core network side, the router nodes comprise core network routers and access network routers, and the user nodes refer to user terminals;
connection topology of each node: the access network router topology can employ a random topology or a ring topology or a mesh topology or a fully connected topology.
5. The P2P streaming media system emulation platform of claim 4, wherein: the attributes of the communication link include link delay data and link bandwidth data.
6. The P2P streaming media system emulation platform of claim 5, wherein: the operation logic refers to a business process and a resource scheduling strategy; the business process comprises a media publishing process and a user access process;
the resource scheduling policy defines data scheduling relationships between user nodes and between a user node and a core network element node, and includes:
and (3) updating the strategy of the neighbor node when the user node joins and exits: when the number of the neighbor nodes of the user node is less than the threshold value, or the service capability of the neighbor nodes is less than the threshold value, or the time from the last time of updating the neighbor nodes exceeds the specified time, the user node updates the neighbor nodes;
data scheduling policy on user node in the playing process: adopting a data scheduling algorithm combining pure pull or push pull;
strategy for increasing the number of channel copies as the amount of access increases: selecting a core network element node which is idle at present, and publishing a channel copy to the core network element node;
policy to reduce the number of channel copies when the amount of access is reduced: selecting a core network element node which is busy at present, and deleting a part of channel copies on the core network element node;
the protocol message comprises definition of each field of the protocol, size of a protocol message packet and an interactive flow of the protocol message;
the streaming media playing mode comprises the following steps:
when the required data fragment arrives, starting normal playing;
and if the required data fragment does not arrive, the playing is suspended.
7. A method for performing performance evaluation on a P2P streaming media system by using the P2P streaming media system simulation platform of claim 6, wherein: firstly, writing all configurations in a static model into a database and a configuration file of a simulation tool as input data, then reading the input data in the database and the configuration file by a program of a P2P streaming media system simulation platform through a dynamic model to perform simulation operation and result data output, and finally evaluating the performance of the P2P streaming media system simulation platform by using the result data; the program of the P2P streaming media system simulation platform is written by using a simulation tool.
8. The method for performance evaluation of a P2P streaming media system according to claim 7, wherein: the method comprises the following steps:
(1) instantiating the static model and the dynamic model, and determining a basic framework and an operation flow of the P2P streaming media system; the instantiation refers to inputting and recording the specific configuration of the three types of nodes into the simulation tool, wherein the 24-hour user request amount in the static model is recorded into a database of the simulation tool, and all the other configurations in the static model and all the configurations of the dynamic model are recorded into a configuration file of the simulation tool;
(2) the program of the P2P streaming media system simulation platform reads the input data in the configuration file and the database, and the control model controls the operation of the program of the P2P streaming media system simulation platform to realize the simulation operation of the P2P streaming media system simulation platform; recording simulation result data in a simulation operation process, wherein the simulation result data comprise timestamps and real-time bandwidth loads on a core network element node and a user node;
(3) calculating to obtain a user experience side evaluation index and a network performance side evaluation index by using the simulation result data obtained in the step (2), wherein the user experience side evaluation index comprises the starting delay and the playing continuity of a user watching channel, and the network performance side evaluation index comprises the bandwidth utilization rate and the neighbor contribution rate;
the starting time delay refers to a time interval from the time when the user node sends a program playing request to the time when the user node starts playing the program, namely, the difference between a timestamp for starting playing of the user node and a timestamp for online of the user node; the lower the index value is, the faster the playing is started, and the shorter the user waits for playing time;
the playing continuity refers to that after the streaming media starts playing, in a time period T, the time for which the streaming media can be continuously played on the user node is T ', and the playing continuity is T'/T, namely the percentage of the total playing time of the user is occupied by the time length of continuous playing; wherein, T is the difference of two timestamps from the beginning of playing to the current time, T' is the sum of all the time segments of continuous playing in the time segment T, and each time segment of continuous playing refers to the difference of two timestamps from playing to pausing; the higher this property the more continuous the viewing;
the bandwidth utilization rate refers to the percentage of the bandwidth used by the core network element node in the available bandwidth and the percentage of the bandwidth used by the user node in the available bandwidth in a sampling time period, that is, the real-time bandwidth load on the core network element node and the user node accounts for the percentage of the configured bandwidth capacity;
the neighbor contribution rate is: the user node provides the proportion of the uploaded neighbor node number to the total neighbor node number, and represents the contribution degree of the user node to the neighbor node;
(4) and completing the performance evaluation of the P2P streaming media system according to the starting time delay, the playing continuity, the bandwidth utilization rate and the neighbor contribution rate.
9. The method for performance evaluation of a P2P streaming media system according to claim 8, wherein: the step (1) specifically comprises the following steps:
(11) instantiating a static model, comprising:
(A1) determining the computing resource data of each node and the communication resource data of each node of the P2P streaming media system, and instantiating a node model;
(A2) determining the number and connection topology conditions of various nodes of the P2P streaming media system, instantiating a topology model, constructing an overall architecture of the P2P streaming media system by combining the node model, and connecting the nodes in the entire P2P streaming media system through the topology architecture according to the configured number and connection topology of the nodes;
(A3) determining link delay data and link bandwidth data of the P2P streaming media system, instantiating a link model, and loading link attributes through configuration about link information in a configuration file in combination with the constructed overall architecture;
(A4) instantiating a user model according to a user arrival model, a user session duration model and a media resource access model; the three models are provided for the user configuration of the P2P streaming media system, the user configures the required parameters according to the requirements, and the P2P streaming media system simulation platform can provide the simulation of the corresponding user behaviors;
(12) instantiating a dynamic model, comprising:
(B1) instantiating a control model according to the operational logic of the P2P streaming media system;
(B2) instantiating a protocol model according to protocol messages of the P2P streaming media system;
(B3) the playback model is instantiated according to the streaming media playback mode of the P2P streaming media system.
10. The method for performance evaluation of a P2P streaming media system according to claim 9, wherein: the simulation tool adopts OMNeT + + or OPNeT.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210157359.0A CN102694799B (en) | 2012-05-18 | 2012-05-18 | P2P (peer-to-peer) stream media system simulation platform and performance evaluation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210157359.0A CN102694799B (en) | 2012-05-18 | 2012-05-18 | P2P (peer-to-peer) stream media system simulation platform and performance evaluation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102694799A true CN102694799A (en) | 2012-09-26 |
| CN102694799B CN102694799B (en) | 2015-05-20 |
Family
ID=46860086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210157359.0A Expired - Fee Related CN102694799B (en) | 2012-05-18 | 2012-05-18 | P2P (peer-to-peer) stream media system simulation platform and performance evaluation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102694799B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103297861A (en) * | 2013-06-05 | 2013-09-11 | 中国科学院自动化研究所 | P2P (peer-to-peer) video-on-demand simulation system on basis of PeerSim |
| CN104918130A (en) * | 2014-03-12 | 2015-09-16 | 腾讯科技（北京）有限公司 | Methods for transmitting and playing multimedia information, devices and system |
| CN105912456A (en) * | 2016-05-10 | 2016-08-31 | 福建师范大学 | User interest migration-based big data simulation and generation method |
| CN106528983A (en) * | 2016-10-26 | 2017-03-22 | 国网安徽省电力公司 | Application system performance evaluation method based on Petri net and analytic hierarchy |
| CN114401421A (en) * | 2022-03-23 | 2022-04-26 | 中国传媒大学 | Method and system for short video transmission strategy based on data driving |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158559A (en) * | 2011-04-18 | 2011-08-17 | 浙江工业大学 | Peer-to-peer (P2P) network-based load balancing method |
-
2012
- 2012-05-18 CN CN201210157359.0A patent/CN102694799B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158559A (en) * | 2011-04-18 | 2011-08-17 | 浙江工业大学 | Peer-to-peer (P2P) network-based load balancing method |
Non-Patent Citations (3)
| Title |
|---|
| SHUANG KAI 等: "Modeling VOIP Service Process Based on Network", 《IEEE》 * |
| 冯侦探 等: "P2P流媒体直播系统自适应邻居节点选择算法", 《西安电子科技大学学报》 * |
| 李振华: "P2P流媒体系统的若干关键属性的优化方案", 《道客巴巴HTTP://WWW.DOC88.COM/P-18564660709.HTML》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103297861A (en) * | 2013-06-05 | 2013-09-11 | 中国科学院自动化研究所 | P2P (peer-to-peer) video-on-demand simulation system on basis of PeerSim |
| CN103297861B (en) * | 2013-06-05 | 2016-08-17 | 中国科学院自动化研究所 | A kind of P2P video request program analogue system based on PeerSim |
| CN104918130A (en) * | 2014-03-12 | 2015-09-16 | 腾讯科技（北京）有限公司 | Methods for transmitting and playing multimedia information, devices and system |
| CN105912456A (en) * | 2016-05-10 | 2016-08-31 | 福建师范大学 | User interest migration-based big data simulation and generation method |
| CN105912456B (en) * | 2016-05-10 | 2019-01-22 | 福建师范大学 | A kind of large data sets simulation generation method based on user interest migration |
| CN106528983A (en) * | 2016-10-26 | 2017-03-22 | 国网安徽省电力公司 | Application system performance evaluation method based on Petri net and analytic hierarchy |
| CN114401421A (en) * | 2022-03-23 | 2022-04-26 | 中国传媒大学 | Method and system for short video transmission strategy based on data driving |
| CN114401421B (en) * | 2022-03-23 | 2022-07-05 | 中国传媒大学 | Method and system for short video transmission strategy based on data driving |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102694799B (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Improving cloud gaming experience through mobile edge computing | |
| Li et al. | Strategy for improving the football teaching quality by AI and metaverse-empowered in mobile internet environment | |
| Shen et al. | Peer-to-peer media streaming: Insights and new developments | |
| CN102694799B (en) | P2P (peer-to-peer) stream media system simulation platform and performance evaluation method | |
| CN105122879B (en) | For observing the mthods, systems and devices of auxiliary Bandwidth Management | |
| Su et al. | Incentive mechanisms in p2p media streaming systems | |
| Zhang et al. | Enhancing the crowdsourced live streaming: A deep reinforcement learning approach | |
| Liu et al. | Deep generative model and its applications in efficient wireless network management: A tutorial and case study | |
| Kryftis et al. | Resource usage prediction models for optimal multimedia content provision | |
| CN101651708A (en) | Topological construction method of P2P streaming media network | |
| Jiang et al. | HD3: Distributed dueling DQN with discrete-continuous hybrid action spaces for live video streaming | |
| Carra et al. | Graph based analysis of mesh overlay streaming systems | |
| Ma et al. | Measurement, modeling and enhancement of BitTorrent-based VoD system | |
| CN117793803A (en) | Multi-user wireless channel resource allocation optimization method based on deep reinforcement learning | |
| Sina et al. | CaR-PLive: Cloud-assisted reinforcement learning based P2P live video streaming: a hybrid approach | |
| Kim et al. | HTTP adaptive streaming scheme based on reinforcement learning with edge computing assistance | |
| CN103368770A (en) | Gateway level topology-based self-adaptive ALM overlay network constructing and maintaining method | |
| Luglio et al. | A Flexible Web Traffic Generator for the dimensioning of a 5G backhaul in NPN | |
| Nguyen et al. | Ossim: A generic simulation framework for overlay streaming | |
| CN109962791A (en) | A kind of stream media system platform and its test method | |
| Lobo et al. | Modeling Video on Demand services taking into account statistical dependences in user behavior | |
| Liotou et al. | The CASPER user-centric approach for advanced service provisioning in mobile networks | |
| Zhanikeev | How variable bitrate video formats can help P2P streaming boost its reliability and scale | |
| Melendi et al. | Modelling and simulation of a real Internet radio service | |
| Liu et al. | Adaptive low-latency peer-to-peer streaming and its application |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150520 Termination date: 20210518 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |