CN101645912A - Internet-based method for building event-driven mechanism programming model - Google Patents
Internet-based method for building event-driven mechanism programming model Download PDFInfo
- Publication number
- CN101645912A CN101645912A CN200810117826A CN200810117826A CN101645912A CN 101645912 A CN101645912 A CN 101645912A CN 200810117826 A CN200810117826 A CN 200810117826A CN 200810117826 A CN200810117826 A CN 200810117826A CN 101645912 A CN101645912 A CN 101645912A
- Authority
- CN
- China
- Prior art keywords
- node
- internet
- data
- event
- multicast
- 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
Images
Abstract
The invention discloses an Internet-based method for building an event-driven mechanism programming model and belongs to the field of Internet application and development. The method comprises: firstly, enabling computing nodes to receive and transmit a request; secondly, establishing a direct and equipotent interactive protocol among the computing nodes and establishing message-based event listening mechanisms at the computing nodes; thirdly, establishing a dynamic cluster among nodes having common needs and transmitting data by using a data multicast mechanism in the dynamic cluster; fourthly, monitoring nodes in the dynamic cluster, and transmitting data of an event occurrence node to other nodes through the data multicast mechanism; and finally, establishing an event-driven programmingmodel according to the type of data transmitted. On the premise of avoiding network cost increase, the method increases network computing resources; and meanwhile, a event-driven mechanism in the programming model allows a network user to acquire concerned event information timely and automatically. Therefore, network service is effective, convenient and humanized.
Description
Technical field
The present invention relates to a kind of method for building up, belong to field of Internet application and development based on the Internet event-driven mechanism programming model.The major technique that wherein comprises has Distributed Calculation, internet communication, remote events driving model, multicasting technology and remote message mechanism etc.
Background technology
Event-driven mechanism has been widely used in the middle of ordinary procedure and the Distributed Programming.As a kind of programming model, event-driven provides a logon mode to monitor potential contingent incident in order to auxiliary system; And after incident took place, timely response was made to incident by the system that utilizes informing mechanism to make that the care incident takes place immediately.If there is not this machine-processed support, developer's programing work amount can increase greatly, and simultaneity factor efficient also can descend.Yet, in the middle of the current internet development environment, do not have such event-driven mechanism.This causes the Internet to develop become complexity and inefficiency.
In the middle of some common distributed systems, event-driven mechanism exists, i.e. the remote events driving mechanism.This mechanism is usually based on remote procedure call (RPC) [1] or RMI (RMI) [2].When realizing the remote events driving mechanism, potential remote events can be recorded on the registrar and with monitoring system and be associated.When on the remote node registered events having taken place, this remote node can initiatively call the monitoring system that is associated with this incident.Monitoring system then excites node relative program of living in to finish response to incident.But this mechanism does not exist in internet, applications.When the developer responds potential the Internet incident at needs, have to take voting method [3] to monitor the generation of incident.Voting method itself is a poor efficiency mode, and system has to pay too much extra cost for this reason; Developer's programing work also can increase; Also be necessary for voting method and design appropriate access cycle, and this cycle must just can receive satisfactory result with specifically using combine.When system was in fluctuation status, this cycle was not fixed, and this also increases added burden access cycle for surveying.Even the analog access cycle can relatively accurately obtain, but because the uncertainty of incident itself, thereby still need consumption of natural resource to reach the potential incident of tracking and after generation, can in time respond.In addition, developer's programing work amount also can increase, and program itself is also succinct inadequately.
It is the feature that meets internet development initial stage applied environment that the current internet system does not adopt event-driven mechanism, is to the adaptation of computing environment at that time.And this adaptation to prevailing circumstances has also caused the obstacle that adopts event-driven mechanism in current internet.Now this is carried out brief explanation.
The first, early stage and current internet is absolute center with Internet Server.Be that early stage the Internet or current internet all are to be absolute center with Internet Server.All computational resources all come from Internet Server; About all serviced device resource of whole features that the Internet calculates institute.This is rational in the internet, applications environment to the absolute mode that relies on of server in early days.At that time, the client computing capability of access internet was limited; Even computing capability is strong, because network bottleneck and safety problem also cause the Internet to calculate other computational resources that are difficult to utilize except server.Event-driven mechanism is to be based upon a plurality of relatively independent computing modules or node all can carry out on the mutual condition by active mode in essence.The computation model at the absolute center of Internet Server itself is inconsistent with the computing environment that the remote events driving mechanism needs.
The second, early stage Internet Server issue resource is mainly to upgrade static resource slowly.Internet Server resource-constrained to be released in early days, its renewal is not frequent yet, far below the frequency of client-access.In this case, new data issue or update event need not to produce and the notice client.In addition, early stage internet, applications is based on static resource.In the middle of the internet, applications based on static resource, except definite events such as data issue, renewal and deletion, there is not more incident to need to follow the tracks of.
The 3rd, early stage and current internet server computational resource is limited.No matter being in early days or the current internet system, is that the pattern at center causes the Internet computational resource can become limited when client is in large scale with the server.In order to deal with potential a large number of users visit with Limited resources, have to reduce the resource consumption of non-main calculating.The remote events driving mechanism not only has pressure to the resource consumption of server own, requires the computation model of client to change simultaneously thereupon.Under the situation that does not have to change in client, it also is understandable also event-driven mechanism being saved during server design.In fact this formed an awkward situation.Saved event-driven mechanism, voting mechanism will increase the pressure of server; And the use event-driven mechanism, server must be paid extra computation pressure equally.Early stage the Internet designer has selected not realize event-driven mechanism after having weighed the pressure that server may be paid under two kinds of situations.This and at that time on the server kind of event few, take place frequently also not relevant.
The 4th, do not exist the network that continues to keep to be connected with current internet in early days.Because potential incident occurrence time is unable to estimate in advance, this network that just requires to keep between the different nodes continuing in distributed environment connects.In case tracked incident takes place, between the node in time ditch pass to response events.Yet the current communication protocol HTTP in the Internet requires the life cycle of each connection limited, thereby guarantees that the limited connection resource of server can be used for dealing with the access request of potential customers' end.Do not continuing to make being implemented on the Internet resources of remote events mechanism be restricted under the network connection situation.Even potential incident takes place, limited Internet resources cause event notification mechanism in time to move.
The 5th, there is not suitable multicast mechanism in early stage and current internet.As the center that the Internet calculates, numerous the clients that have of Internet Server incident are paid close attention to usually.Under such prerequisite, require Internet Server have one efficiently multicast mechanism to alleviate Internet Server because of the pressure that brings of notice client.Yet in the current internet system, this multicast mechanism is non-existent; In case there is an event-driven mechanism, server has to pay the client that computational resource notifies all that incident is paid close attention to, thereby causes calculating pressure excessive.This also is that current internet does not have one of reason of event-driven mechanism.
The 6th, pure thin-client pattern causes client can't participate in the Internet and calculates.Another main cause that causes the current internet system not have the existence of incident driving mechanism is pure thin-client pattern.So-called pure thin-client pattern refers to client under any circumstance only to send the form access server of asking and accepting to respond.And this form is not the participation that the Internet is calculated in essence, just the initiation that the Internet is calculated.All calculate to have fully under the client-requested and carry out on Internet Server.Event-driven mechanism is to be based upon on the basis that various computing node fellowship is calculated.But calculate because client can not participate in the Internet, between client and server and all can not produce based on the calculating of cooperating between the client, so event-driven mechanism all can't be set up in early days with in the current internet.
Summary of the invention
The object of the present invention is to provide a kind of method for building up based on the Internet event-driven mechanism programming model, it is set up based on remote events driving mechanism and programming model for the developer in internet, applications, simplify the Internet application and development cost, and make contributions for setting up more efficient internet, applications.
Technical scheme of the present invention is:
A kind of method for building up based on the Internet event-driven mechanism programming model the steps include:
1) adopt TCP/IP peer protocol pattern to make computing node have the ability of the request of reception and transmission request;
2) between computing node, set up the directly interaction protocol of equity;
3) on computing node, set up message based action listener mechanism, be used to receive data;
4) set up dynamic cluster, and in dynamic cluster, adopt data multicast mechanism to carry out transfer of data; Described dynamic cluster is the computational resource cooperation Share Model that forms in some moment based on the computing node of joint demand;
5) node of monitoring in the described dynamic cluster is transferred to other node with the data of event occurrence node by described data multicast mechanism;
6), set up event-driven programming model according to the data type of transmission in conjunction with software engineering technology.
Described computing node comprises the client and server end.
The method for building up of the interaction protocol of direct equity is between the described computing node:
1) sets up network and connect monitoring mechanism;
2) pass through the direct connection that IPv6 or NAT crossing technology realize arbitrary node on the Internet;
3), set up the different mutual agreements of client and server end equity according to concrete internet, applications feature.
Utilize asynchronous programming model to finish the described message based action listener mechanism of setting up in the described method in conjunction with ICP/IP protocol.
The method of work of described message based action listener mechanism is: after obtaining the related remote event message, start the local event driving mechanism remote events is converted to local incident, drive local related system remote events is responded.
Described related remote incident comprises: message remote events, file remote events, byte remote events and object remote events.
Described data multicast mechanism is that the topological structure that utilizes the BT agreement to form carries out the multicast of lightweight data, and its method is:
1) adopt the state of computing node in the tracker monitoring dynamic cluster of BT agreement, described node state comprises: whether node is in the cluster, node whether to certain incident pay close attention to, whether node be whether super node, node have obtained certain incident;
2) for being in node in the cluster, described tracker equilibrium distributes the multicast task, and its method is: at first for the transmission data unique identification is set; Described then tracker with the super node determined as kind of a child node, by concurrent connection to other node transmission data; The sign of the data that the node inspection receives if repeat, then stops to be transferred to the concurrent connected node of this node; Otherwise the concurrent connected node that is transferred to this node continues multicast.
Sign has the node of process once on the described transmission data, when the node that current arrival node and lightweight data process occur repeats, then stops transmission; Otherwise, continue multicast process.
Described data type comprises: message, file, byte and object.
Described event-driven programming model comprises: message transmission and dependent event, file transmission and dependent event, byte transmission and dependent event and object send and dependent event.
Method flow of the present invention as shown in Figure 2, its main contents are:
The first step for remedying the computational resource deficiency, is set up the internet environment that all nodes all can participate in calculating.The prerequisite of setting up the remote events driving mechanism is to solve current internet to the absolute problem that relies on of server.This operational mode makes computational resource seriously be subject to Internet Server.For covering the shortage, the mode that increases the server computational resource separately needs, but is not effective way.When facing the pressure of concurrent request of potential a large amount of clients and heavyweight transfer of data, it is not enough that the computational resource that aforesaid way provides remains.Solve the problem of the Internet computational resource deficiency fully, only take that all are in node in the internet environment and all add the Internet and calculate, and utilize they self computational resource.For finishing above-mentioned work, need the work of following step.
(1) changes pure thin-client server mode.The pure thin-client pattern of current internet is to cause the main cause of the Internet computational resource deficiency.The solution of this problem is the key that increases the Internet computational resource.In early days the internet development stage, server resource is unique source of the Internet computational resource.But along with computing equipment and development of Communication Technique, the computing node that is on the Internet has all possessed available ability on computing capability and bandwidth.If these computing nodes still occur in internet, applications with pure thin-client pattern, only can produce consumption to Internet resources.Along with popularizing of internet, applications, this pure thin-client can get more and more, also can be more and more fierce to the contention of Internet resources.If can not change this pattern, Internet development can only rely on server end powerful computing ability to occur.This meaned high cost investment concerning most of internet, applications, be unpractical solution.
(2) change client and server end interaction protocol.The interaction protocol of active client and server end is with the client request of sending and awaits a response, and server end receives request and produce and respond to basic working modes.This pattern is born all calculation costs by server; Client does not participate in calculating.The present invention has changed this agreement, transfers to client calculating by server end, to alleviate the server computation burden as far as possible.The main principle of this change is that the formal transformation that request is responded is an ad-hoc mode.Realize this pattern, need give client and give the ability of server end with the request of sending to receive the ability of asking.As ICP/IP protocol, itself is exactly a peer protocol.But since based on the internet, applications agreement HTTP of TCP/IP in order to adapt to the concrete environment of early stage the Internet, ICP/IP protocol has been made into client pattern has been responded in the simple request of server end.The present invention has become this pattern into ad-hoc mode by means of current computing equipment and development of Communication Technique.To solve three major issues yet reach such purpose on the internet, just can obtain being fit to the peer protocol of internet, applications.The first, set up network and connect monitoring mechanism.Http protocol only possesses network monitoring mechanism at server end, and does not have in client.This is to have only the server end could be accessed and reason that client is initiatively visited.Among the present invention, must give the ability that the two possesses network monitoring.The second, realize the connection of arbitrary node on the Internet.The Internet is owing to address space and safety factor, and not all node can both be set up direct network and connect.This obstacle can pass through IPv6[11] or NAT crossing technology [12] solve.The 3rd, according to concrete internet, applications feature, set up the different mutual agreements of client and server end equity.Because the Internet complexity of calculation, applications client of different nature and server end interactive mode vary.In order to deal with different scenes, the present invention is after the connection of having set up the client and server end, the basic transmission means of message, file, byte and object of equity also is provided between the two, and the developer is based on this, be customized to be fit to concrete use interaction protocol.
(3) set up interaction protocol between the client.Except setting up novel interaction protocol, also can produce cooperation between the client at the client and server end.In the current internet environment, do not exist any mutual between the client.It is assisted indirectly by server fully alternately and finishes.This has also increased the calculating pressure of server.The present invention is by setting up direct interaction mechanism between client, thereby calculates primary condition is provided for forming dynamic cluster between the client.This is vital for alleviating the server end calculating pressure and increasing whole the Internet computational resource.Realize this target, need to solve monitoring mechanism, connection and combination equally and specifically use three problems.In the middle of these three problems, setting up interaction mechanism between client, to bring in meeting than client and server more difficult.Its main cause is the influence that client can be limited by limited address space in the Internet and safety factor more.If can being higher than at the client and server end, use NAT crossing technology, its complexity connect.This is because each server possesses addressable internet address usually, and does not possess the internet access address or the possibility that is in the NAT protection wants high with client.
(4) set up interaction protocol between the server end.Similar with the active client interactive mode, assisting by client alternately between the server end finished.It is pointed out that current is that mutual alternately or between the server end between the client all being finished in artificial the participation indirectly, is not that the dynamic cluster of mentioning by the present invention is finished between above-mentioned node automatically.This calculating based on artificial participation can only form the mutual of lightweight poor efficiency, is unable to cope with for heavyweight high efficiency complicated calculations.For poor efficiency interaction problems between the settlement server, need between these servers, set up directly mutual equally.For this reason, monitoring mechanism, connection and concrete application characteristic also need to consider.Comparatively speaking, the solution of the problems referred to above is just much easier between server.This mainly is that Internet Server all possesses other computing nodes visits of independently internet address confession usually, and server performance generally will be better than numerous clients greatly simultaneously.In fact, after connecting, it is the most difficult taking concrete agreement at different application.The present invention has set up the basis for more effective the finishing of this work.
(5), set up corresponding interaction protocol at concrete application.Real actual interaction protocol must and specifically be used and combine closely, and takes related algorithm or solution according to the feature of concrete application.Internet, applications roughly can be divided three classes according to the feature of transmission data, and promptly lightweight is used [4], used [5] and use [6] with sequential associated weight level with the irrelevant heavyweight of sequential.Lightweight is used and is suitable for general instant messaging and business system; Be suitable for file-sharing, Email, BBS and common Web application etc. with the irrelevant heavyweight application of sequential; Being suitable for the audio frequency and video stream medium data with the application of sequential associated weight level is main application system.The character of different application all is differentiated to the requirement and the mode of operation of computational resource.But because essential change has taken place the distributed basic structure that the present invention relied on, thereby cause above-mentioned different application under abundanter computational resource support, to work.Detailed algorithm can be with reference to pertinent literature and my other inventions.
In second step, set up message based action listener mechanism.Mutual restriction has caused the remote events driving mechanism to be achieved in the current internet system between the in short supply and computing node of computational resource.By the change to internet, applications layer basic structure, above-mentioned two subject matters have obtained solution, and promptly the Internet event-driven mechanism is to be based upon between the abundant and computing node of computational resource on the free mutual basis.Under this prerequisite, just can on each computing node, set up action listener mechanism.This mechanism is that lightweight data (message) are waited for interface in essence.This interface can be finished in conjunction with ICP/IP protocol by utilizing asynchronous programming model [7].It should be noted that after obtaining the related remote event message, need to start the local event driving mechanism remote events is converted to local incident, drive local related system remote events is responded.Current main operating system and developing instrument are all supported the local event driving mechanism, can finish this work with reference to relevant material.
In the 3rd step, set up high efficient and reliable multicast mechanism based on the lightweight data.In the computing environment that the present invention relied on, use all be dynamic cluster finish (the present invention has adopted in complex networks system [8] that data are duplicated, netted multicast mechanism [5,6], dynamically screen node [5] and set up dynamic cluster based on the routing algorithm means such as [9] of Complex Networks Theory).So-called dynamic cluster is based on the computational resource cooperation Share Model of the computing node of joint demand in some moment formation.This model inside reaches the purpose that improves the system-computed performance by sharing of computational resource.It is different fully with the cluster under stable environment to set up the method for dynamic cluster.The drift net agreement is an important solutions wherein, and it is summarized as follows simply:
Foundation is based on the routing algorithm of community network [13].This routing plan is the resource finding algorithm that is based upon on community network or complex network [13] basis.Meet the topological structure that Internet resources distribute rule because community network is computing node in essence by what interconnected user on the network's social property was set up, utilize this network just can finish effective searching rightly computational resource.When utilizing community network to carry out route search, relate generally to structuring and destructuring route technology in conjunction with problem.The outstanding feature of structuring route technology is to obtain route results in determining finite time to any resource (rare or popular), but such result is based upon under the strict constraints.And this condition current internet environment is difficult to be guaranteed.The destructuring route technology is based upon on the topological structure (being community network) that the Internet forms naturally, the internet environment that exists is not had any mandatory requirement.Will guarantee to obtain Search Results fast on such basis, necessarily requiring has deep understanding just can reach to concrete application and corresponding topological structure.But even so, unstructured searching is still not enough in the assurance of route fairness, and is also difficult to sharing of route pressure simultaneously.In view of the foregoing, can be the appropriate combination of said two devices.Because have group in the community network, group is representing metastable computing environment, approaches the desired condition of structuring route.Like this, can take in whole the Internet, to carry out the destructuring route, and in group, carry out the structuring route.By said method, routing algorithm has just been accomplished the popular resource of effective searching and has been guaranteed route justice to scarce resource.
Set up the event-driven mechanism in the dynamic cluster.One of basis that dynamic cluster forms is exactly to be prerequisite each other alternately with the event-driven.Be that a node request connects, initiates transfer of data and closes connection or the like basic act and all finish by event-driven mechanism.This and traditional the Internet are completely different in the mode of ballot.And after the basis of setting up dynamic cluster, this event-driven mechanism has just formed basic framework.In conjunction with lightweight multicast mechanism efficiently, just can form a message based event-driven mechanism again.This mechanism can adapt to the feature of the Internet dynamic change, and the assurance event-driven mechanism is worked under the reliable extensive environment of efficient stable.The realization of this mechanism has improved the interactive efficiency between the Internet computing node, has avoided voting mechanism to the pressure of system's increase and to the system change torpid reaction.Event-driven has guaranteed that concern person can need not pay extra cost and vote before incident takes place; Simultaneously, after incident takes place, can react at once.Because this cluster is not stable, but is in the middle of the continuous variation.This variation comprises the adding of computing node and leaves, and also comprises the change of state of computing node own and behavior, also comprises the change of computing environment and the variation of computing node character.For the such special distribution formula computing environment of dynamic cluster, these variations all are at first to take place in some parts; The variation of these local states may influence the performance of overall calculation.For this reason, a plurality of nodes all can be paid close attention to certain dynamic change, and in time react under the situation that changes generation.So just formed demand to event-driven mechanism.And the local state variation will be reflected to the overall situation, needs a reliable multicast mechanism to finish.This also is the necessary demand of the Internet remote events driving mechanism.This multicast mechanism need be finished following work.
(1) uses lightweight data representation incident and correlation behavior.The key of remote events driving mechanism is that incident after certain local generation, can in time make local state be converted into wider or even global state.The data of this state of ordinary representation are little.Even state is too much in the system, being illustrated in the real system of its correspondence also can be used the lightweight data description.So in the multicast system that is thought of as the event-driven service, only need to consider that the lightweight data get final product.This has alleviated the design difficulty of multicast system.
(2) problem that will consider of lightweight multicast system.At first be to disperse multicast pressure.The node scale that faces when the Internet multicast system is big, even the multicast of lightweight data takes the notification strategy of simple one-to-many also can cause calculating pressure to the event notice node, so must consider that suitable means are to alleviate this pressure.Secondly, consider adaptation to dynamic environment.Another feature of internet, applications is exactly that the behavioral characteristics of computing node is obvious.Except being the server (peer server) of the free dedicate resources of system, all the other nodes all participate in the Internet and calculate under user's control.Satisfied its requirement or when nonsensical, just can withdraw from this calculating at any time when the user thinks that the Internet calculates it.On the other hand, a computing node also can add in the Internet computational process at any time.The dynamic that these two factors all can cause the Internet to calculate.Like this equally for multicast system.In the process that multicast is carried out, the node that is in the multicast system is not in stable state.These behavioral characteristics can have a strong impact on the reliability of multicast.Once more, consider control to instantaneity.The instantaneity that key character of event driven system is exactly the requirement event notice.When the multicast system scale was big, the assurance of instantaneity will go wrong.The instantaneity that guarantees extensive multicast system is the important need that guarantees that the remote events driving mechanism is set up.At last, to the assurance of reliability.Reliability mainly refers to sequence consensus and the fault-tolerance of data in the dynamic environment transmission course in the lightweight multicast system.When a plurality of incidents when taking place concomitantly, multicast system must guarantee the incident original time that coincident event takes place when being transferred to each node order.For large-scale distributed system, the order of incident takes place chaotic easily in multicast process.Fault-tolerance must be accomplished in the frequent distributed system of dynamic change, can not be guaranteed to cover all interdependent nodes to prevent the incident multicast.
(3) utilize the BT agreement to realize lightweight data multicast system.At the problems referred to above, the present invention has adopted based on the multicast system that improves the BT agreement.The BT agreement is a heavyweight data multicast protocol in essence.Why consideration utilizes the BT agreement, mainly is that its running environment is consistent with environment of the present invention; The BT agreement is worth using for reference for the accommodation mode of dynamic environment.Only pay close attention to the feature of lightweight data at event driven system, the present invention must make improvements, and makes it be more suitable for the multicast of lightweight data.Fig. 1 has illustrated and has utilized BT to realize lightweight data multicast basic mode.
A. use tracker monitoring dynamic cluster node state.Tracker in the BT agreement is the server that is used for collecting all node states in the potential dynamic cluster specially.In the present invention, this server can be realized by arranging central server; In the present invention, the maximum characteristics of this central server be can be initiatively with system in computing node get in touch, rather than always be in the status that passive wait computing node is reported oneself state.These characteristics cause tracker can become an Event triggered node based on BT.When needs and the contact of some computing nodes, can initiatively send data or correlation behavior, and need not passive wait.Such as, tracker is used to evaluate super node in the present invention, and in view of the above with these nodes initial as the multicast of lightweight data; This need of work has tracker to get in touch super node in mode initiatively.For being used for carrying out event driven lightweight multicast system, need the state of monitoring to comprise whether node is in the cluster, node whether to certain incident pay close attention to, whether node be whether super node [10], node have obtained this incident etc.These states all can play a key effect for the quality of incident multicast system.After node state is held in to system, could rationally share, improve multicast efficient and reliability lightweight multicast role pressure.
B. the topological structure that utilizes the BT agreement to form carries out the lightweight multicast.The BT agreement is used for carrying out heavyweight transfer of data or multicast.In order to guarantee the multicast of heavyweight data efficient at high dynamic environment, the BT agreement has been taked a series of schemes, as node selection and data slot selection etc.The essence of BT agreement is based on the purpose that reciprocal policy calculation resource-sharing reaches the heavyweight multicast as noted earlier.No matter carry out node selection or data slot and select or the principle of reciprocity, its free-revving engine is the coupling that realizes application layer topological structure and network layer topology structure.So-called application layer refers to the data interaction layer of setting up according to the concrete needs of using; Network layer refers to and is independent of the based on network data interaction layer of any concrete application.Application layer is based upon on the network layer; The data interaction of any application layer finally all must be transmitted on network layer.But the mutual topological structure that forms is directly related with concrete application scenario between the application layer computing node, and between the route transmitted of network enabled topological structure without any contacting directly.When application layer topological structure and network layer topology structure seriously do not match, can make the data interaction inefficiency on the application layer.The height of application layer interactive performance depends critically upon the topological structure of itself and the matching degree of network layer topology structure.The main effect of BT agreement just is to have realized approximate match between them by a series of means.From beginning the effect of back a period of time alternately, can form an application layer of with network layer topology structure proximate mating between the node of all participation heavyweight transfer of data by the BT agreement.This coupling has guaranteed the efficient of heavyweight data multicasts just.Running environment of the present invention also is based on large-scale internet, has formed mutual based on concrete application scenario between the node, i.e. application layer.But having only by the auxiliary of BT agreement, this application layer topological structure could form approximate match with the network layer topology structure.Corresponding therewith, the internet system that the present invention set up is based on the multicast of heavyweight data, and is the multicast solution with the BT agreement, and this provides condition for the coupling that forms with the network layer topology structure.For the multicast of lightweight data, can utilize the application layer topological structure that in heavyweight data multicast process, forms fully.Because effective adjustment of BT agreement has been passed through in the formation of this topological structure, also can keep high efficiency when carrying out the lightweight multicast.
C. the concurrent cooperation number of nodes based on the BT agreement is not imposed any restrictions.Traditional BT agreement during to mutual between the node concurrent quantity a restriction is arranged, be generally four.This restriction is correct for the ordinary node on most of the Internets.But,, but be not suitable selection for internet system of the present invention.Said as the front, the internet system that the present invention set up not only comprises common computing node, more comprises traditional server; These nodes are the main candidates that form super node.These servers have abundant computational resource usually, can concurrent foundation far surpass four multicast connection.Therefore, break this restriction and can make that the node resource of BT agreement operation is made full use of as far as possible, improved heavyweight data multicast efficient.
D. tracker is determined super node according to monitoring.In the BT agreement, the responsibility of tracker is that the node to new adding system just returns part of nodes movable in dynamic cluster, makes new node carry out mutual with these nodes and adds cluster, thereby finish the multicast of heavyweight data.In the present invention, the tracker major responsibility is by provide potential cooperative nodes to be transformed into the balanced multicast task of distributing of the existing node of cluster for new node.In common BT agreement, tracker can return [5] node that (is generally 80) in right amount at random and give the node that is participating in the multicast of heavyweight data.In the present invention, adopt in whole node and determine super node, allow these nodes need at first obtain the lightweight data of multicast according to monitoring.Generally speaking, these super nodes possess more concurrent connection, can make the lightweight data be transferred to other interdependent nodes with high as far as possible efficient by these concurrent connections.Find out the importance of determining super node thus.Determining of super node can be by a simple method realization, and the node time of enlivening of promptly writing down on tracker is weighed: the time of enlivening is long, and the possibility that then becomes super node is big; Otherwise, be exactly ordinary node.Usually the enliven time of super node in dynamic cluster substantially exceeds ordinary node, therefore easily according to this numerical value identification super node.In addition, even super node identification has error, also can find some nodes relatively reliably according to the time of enlivening; They can be known as time super node.Why can utilize some times super node, also be because lightweight data multicast duration burden short, that this node is caused is little.Super node and the secondary super node selected according to mentioned above principle can become kind of a child node fast.Afterwards, these nodes can be by current concurrent connection transmission lightweight data.
E. change the node selfishness and be contribution property.In the BT agreement, each node all is assumed to be in the mode of selfishness and works, and is promptly obtaining will to leave system after the required heavyweight data; Simultaneously with other node reciprocal processes in also just satisfying just understand under the oneself requirement prerequisite and the other side mutual; Otherwise the cooperation meeting is interrupted causing logging off or seeking other nodes and carry out alternately.Multicast is correct for the heavyweight data in such design.But for giving the little lightweight data of system pressure, just there is no need.In the lightweight multicast system, for finishing the multicast task as early as possible, the concurrent connection transmission lightweight data that the node of each acquisition multi-case data should initiatively have by self.Be that super node or ordinary node are all worked in contribution property mode when the multicast of lightweight data.The concurrent connection number that common ordinary node has will be far smaller than super node, and this is couplings with its bandwidth resources, can reach it and can bear maximum pressure.Multicast pressure just balancedly is distributed on each node like this, has guaranteed the instantaneity of multicast simultaneously.Because multicast carries out in astable non-rule topology structure, system need not to safeguarding that topological structure pays a price, and the behavioral characteristics of node is dropped to minimum for the influence of multicast quality.
F. avoid the transmission of lightweight data redundancy by giving the lightweight data with unique identification.Because the lightweight data multicast among the present invention does not have the topological structure of rule, this causes the redundant transmission of data easily, and promptly a node is received repeatedly repeating data.Give the lightweight data with unique identification, when a node is received data, can check this sign.If repeat, then stop to be transferred to its concurrent connected node; Otherwise the concurrent connected node that then is transferred to it continues multicast.
G. the path by lightweight data process identifies and avoids occurring loop.Because the lightweight multicast among the present invention does not have the topological structure of rule, in data transmission procedure, might form loop equally, cause nonsensical transfer of data.For this reason, can on the lightweight data, identify the node of process once.When the node that current arrival node and lightweight data process occur repeats, then stop transmission; Otherwise, continue multicast process.
H. overcome the dynamic of the Internet by dynamic cluster.Internet environment is because the people to the direct control of computing node, causes system dynamics obvious.Extreme behavioral characteristics can stop the carrying out of any calculating.In fact, in internet environment, the behavioral characteristics that the user causes not is random, but determined by quality and the character of concrete application own.For whole internet environment, behavioral characteristics is absolute; But for concrete an application or concrete some periods of using, behavioral characteristics is relative, even can show invariant feature.The concrete application that is established as of dynamic cluster provides metastable computing environment in the dynamic internet.If in this environment, carry out the multicast of lightweight data, as a rule can be subjected to influence dynamically can be reduced to minimum.
I. guarantee the order of lightweight data in multicast process.In some applications, the order of lightweight data is significant.Any one node must respond in proper order according to this just correct result.At least be exactly like this for event-driven mechanism.For the lightweight multicast, guarantee that the order of data in transmission course is the basis of event-driven mechanism true(-)running.On the dynamic the Internet of height, as safeguarding that tracker has played coordinative role to the order between the data.When sending the lightweight data on the node and carrying out multicast, at first to send request to tracker; Tracker sorts according to tracker local zone time sequencing to all nodes that need send data in the cluster; Multicast is carried out successively according to this sequencing, and promptly after the node data of confirming to come the front sent successfully in whole cluster, the node data that comes thereafter sent again.But such way can cause whole system excessively synchronously and decreased performance.Also can take to carry out serial number, carry out asynchronous transmission then being sorted data.When each node was handled the data that obtain, if find that numbering is discontinuous, then can judge had data not receive, can get in touch with data source and obtain related data.This method can be so that the lightweight multicast guarantees reliability under high efficiency prerequisite.
In the 4th step, set up the remote events driving mechanism under the internet environment.On the basis of above-mentioned work, the present invention proposes a remote events driving mechanism based on internet environment.This mechanism can adapt to the feature of the Internet dynamic change, and the assurance event-driven mechanism is worked under the reliable extensive environment of efficient stable.The realization of this mechanism has improved the interactive efficiency between the Internet computing node, has avoided voting mechanism to the pressure of system's increase and to the system change torpid reaction.In voting mechanism, an important parameter is exactly a polling cycle.General this cycle can formulate according to event property in the concrete application and incident concern person's demand.When the user need pay close attention to institute might incident take place the time because the source that is decided by firing event of incident is difficult to prior estimation, so the formulation of polling cycle be difficult to reach accurately, thereby cause polling cycle and incident pests occurrence rule inconsistent.When polling cycle during greater than the incident occurrence frequency, concern person will be slow to reacting of incident; When polling cycle during less than the incident occurrence frequency, remote events generation node and concern person's node drop into more resources with regard to needs and vote, and have additionally increased the pressure of system.Event-driven has guaranteed that concern person can need not pay extra cost and vote before incident takes place; Simultaneously, after incident takes place, can react at once.This shows that the remote events driving mechanism becomes the basis that the reflection the Internet changes under the internet environment.The Internet is a system that constantly changes in essence, but this variation can't be reflected to the correlation computations node immediately under the support that does not have the remote events driving mechanism.In order to reduce computation burden, during current internet is mainly used in addition the refusal these variations are followed the tracks of.The present invention has solved this problem effectively.
In the 5th step, the remote events of setting up under the internet environment drives programming model.Realizing on the Internet after the remote events driving mechanism that the present invention has proposed to drive programming model based on the remote events of internet environment in conjunction with software engineering technology.This model shows as cover application programming interfaces (API) concerning the developer.By using these interfaces, the user can finish monitoring and the response to various possibility incidents on the internet.This socket comprises that message transmission and dependent event, file transmission and dependent event, byte transmission and dependent event and object send and dependent event.Simultaneously, also realized the corresponding asynchronous interface [7] of these interfaces.These cover application programming interfaces have comprised elementary event programming model in the Internet calculating.In internet, applications, calculate all relevant remote events with the Internet and all show with data transmittal and routing form.Therefore, the incident that relates among the present invention all with data transmission interface tight association.The transmission that the Internet calculates for data can be divided into four types according to the character of data, i.e. message, file, byte and object.These types all are the elementary events that is independent of concrete application.When using this model, the developer need send data according to application programming interfaces provided by the invention, and the event type definition that provides is provided or derives and specifically use relevant incident.This wherein object send and dependent event provides remote events defining mode flexibly; The developer can define various incidents based on the object transmission according to self-demand.In the actual exploitation,, just can derive any remote events as long as inherit this abstract object.Table 1 has been listed the Internet remote events and has been driven main DLL (dynamic link library).
Table 1. the Internet remote events drives main DLL (dynamic link library)
| Sync cap | Asynchronous interface | Dependent event | Explanation |
| ??bool ??SendMessageByPeerID(string ??peerID，string?message) | IAsyncResult BeginSendMessageByPeerID(string?peerID， string?message，AsyncCallback?callback， object?state) bool EndSendMessageByPeerID(IAsyncResult ar) | ??OnMessageSent ??OnMessageReceived | Message transmission interface and dependent event thereof |
| ??bool ??SendFileByPeerID(string ??peerID，string?fileName) | IAsyncResult BeginSendFileByPeerID(string?peerID， string?fileName，AsyncCallback?callback， object?state) bool?EndSendFileByPeerID(IAsyncResult ar) | ??OnFileSent ??OnFileReceived | File transmission interface and dependent event thereof |
| ??bool ??SendByteByPeerID(string ??peerID，byte[]buffer) | IAsyncResult BeginSendByteByPeerID(string?peerID， byte[]buffer，AsyncCallback?callback， object?state) bool?EndSendByteByPeerID(IAsyncResult ar) | ??OnByteSent ??OnByteReceived | Byte transmission interface and dependent event thereof |
| ??bool ??SendObjectByPeerID(string ??peerID，object ??peerDataObject) | IAsyncResult BeginSendObjectByPeerID(string?peerID， object?peerDataObject，AsyncCallback callback，object?state) bool EndSendObjectByPeerID(IAsyncResult?ar) | ??OnObjectSent ??OnObjectReceived | Object transmission interface and dependent event thereof |
Good effect of the present invention is:
The present invention is under the prerequisite that does not increase network cost, improved the network calculations resource, simultaneously the event-driven mechanism of programming model of the present invention inside can make the network user in time obtain the event information of being concerned about automatically, makes that the network service is more efficient, convenient, hommization.
Description of drawings
Fig. 1. based on the lightweight data multicast of BT topological structure;
Fig. 2. method flow diagram of the present invention.
Embodiment
Application of the present invention almost can cover all internet arenas.At this, implement situation of the present invention during some are mainly used and illustrated.Utilize the present invention to implement in a plurality of below application.
The first, enrich internet, applications.What is called enrich internet, applications be for solve traditional internet, applications show too simple, with the tangible problem of desktop application gap.One of internet, applications and desktop application difference are exactly not have event-driven mechanism.Have to way by constantly refreshing (to server requests) of user obtains the internet data that upgrades from the website.Application based on desktop then is not like this.Because current desktop operating system all possesses event-driven mechanism, when any incident of user's care took place, the user can obtain event notice under the situation of not carrying out any operation, the line correlation of going forward side by side reaction.The current Internet technology of enriching attempts to replace the user manually to server requests by the bottom voting mechanism, gives the user sensation with case mechanism.But the discussion by the front can obtain conclusion, compares with real event-driven mechanism, and voting mechanism is not an effective solution.Utilize technology of the present invention, can properly settle the problem of realization event driving on the internet.The developer can describe any remote events out by programming model provided by the invention, and this model is combined with concrete internet, applications, forms at concrete application event driving mechanism.
The second, e-mail system.E-mail system also is that current internet mainly one of is used.The transmission of Email with collect itself and just represent different incidents.These incidents be that the user is concerned about.E-mail system based on the remote events driving mechanism can in time point out the user to receive Email or send Email success.The current internet e-mail system does not have the support of remote events driving mechanism, when the user receives Email, can not in time be notified.Have only when user's active access mail server, just can learn current mail state.Some e-mail systems can only use voting mechanism to finish this work.Utilize technology of the present invention, can send mail and collect and describe out by the object transmission interface; No matter be that the user sends or gets the mail and can both in time be notified.
The 3rd, instantaneous communication system.Instant messaging is a kind of high-frequency remote events mechanism.The instant messaging mechanism of current internet does not have the support of event-driven mechanism, can only rely on the acquisition that voting mechanism forms the simulation instant message.Utilize the present invention, can effectively realize sending message in the instantaneous communication system and receiving message.When realizing this target, the DLL (dynamic link library) among the present invention can be utilized, and almost without any need for change.
The 4th, e-commerce system.Demand to the remote events driving mechanism in the e-commerce system is a lot.Some e-commerce system Data Update is frequent, and the user also pays close attention to these renewals, such as stock system.Business data changes frequently in the stock system, and these data all are that the user pays close attention to.The current voting mechanism that can only rely on is realized above-mentioned functions.Utilize technology of the present invention, stock change data can effectively obtain rapidly.Here the problem that also has simultaneously multicast.Usually stock change data have a plurality of users to pay close attention to; Simple event notification mechanism efficient is low, can not guarantee instantaneity.Because the present invention has unique multicasting technology, make stock certificate data in numerous users, effectively spread.For other e-commerce systems, also there is similar problem.This problem embodies a concentrated reflection of in the distributed transaction management [3].As commerce services of numerous user's fellowships, can relate to change to same business data.If there is not effective transaction management mechanism, can cause the inconsistent of business data, and the final confusion that causes, cause commercial loss.When numerous users participate in a commerce services,, require effective event notification mechanism and multicast mechanism in order to keep consistency.Technology of the present invention has just in time satisfied this demand.
The 5th, network game system.There is the mutual of mutual and user and system between a large number of users in the middle of the online game.These can be abstracted into various incident alternately.But the current network games system also is to continue to use traditional voting mechanism to carry out event-driven and relevant multicast mechanism.This has increased system load and has reduced efficient.Utilize technology of the present invention, between the game user and the incident between game user and the system all carry out with effective and efficient manner.When the incident of numerous users' care took place, multicasting technology of the present invention also can play effective support effect.
The 6th, information system.The notable feature of information system is exactly that the frequent and user of data change is to changing sensitivity.Yet because the common customer volume of information system is big, for fear of system being caused excessive pressure, these systems even the effective multicast mechanism that does not have realization event mechanism and be closely related with case mechanism.From here as can be seen, the backward state of current internet technology.Utilize the present invention, any change in the information system can be at once by user's perception; Simultaneously, also can be notified the associated user immediately by the incident of a large number of users perception for one by the multicast mode.
The 7th, the announcement board system.In essence, the announcement board system is consistent with the internet, applications character that generally is issued as feature with data.Some differences are that announcement board system interaction frequency is than general internet delivery system height.Similar with the general internet delivery system, the announcement board system also needs to finish by event-driven mechanism and relevant multicast mechanism the notice and the transmission of instant change data.Based on the reason same with information system, current announcement board system does not have above-mentioned mechanism that system is assisted yet.The event-driven mechanism that also needs alternately on announcement board is notified the associated user; This function does not realize in current announcement board system yet.The technology that utilization the present invention relates to, the problems referred to above can be resolved.
The 8th, conference system.The Internet conference system is that a plurality of users participate in interactive system in essence.Have different user to need distributing data in the middle of the meeting, the issue of these data can show by the form of incident.Use the present invention, can make conference process efficiently smooth, the variation of participant's perception immediately conferencing data.Simultaneously, some conference system needs the support of heavyweight data.The transmission of these heavyweight data can utilize based on the Streaming Media transmission mechanism [6] of BT agreement and finish.The topological structure that this mechanism constitutes has booster action to multicast mechanism of the present invention, and the lightweight multicast that the present invention is relied on shows higher quality.
Reference
【1】Andrew?D.Birrell，Bruce?Jay?Nelson；Implementing?Remote?Procedure?Calls；ACMTransactions?on?Computer?Systems(TOCS)，Volume?2Issue?1，1984
【2】Jason?Massen，Rob?van?Nieuwpoort，Ronald?Veldema，Henri?E.Bal，Aske?Plaat；AnEfficient?Implementation?of?Java’s?Remote?Method?Invocation；PPoPP’99：Proceedings?of?theSeventh?ACM?SIGPLAN?Symposium?on?Principles?and?Practice?of?Parallel?Programming，1999
【3】George?Coulouris，Jean?Dollimore，Tim?Kindberg；Distributed?Systems?Concepts?andDesign；Pearson?Education，Ltd；2005
【4】Obraczka，K.；Multicast?Transport?Protocols：a?Survey?and?Taxonomy，Communication?Magazine，IEEE，Volume?36，Issue?1，Jan.1998Page(s)：94-102
【5】Cohen?B.；Incentives?Build?Robustness?in?BitTorrent，in?Workshop?on?Economics?ofPeer-to-Peer?Systems，Berkeley?USA，May?2003
【6】Vlavianos?Aggelos，et?al.BiToS：Enhancing?BitTorrent?for?Supporting?StreamingApplications，INFOCOM?2006，25
th?IEEE?International?Conference?on?ComputerCommunications?Proceedings，April?2006，Page(s)：1-6
【7】Microsoft?Asynchronous?Programmming?Overview；
http://msdn.microsoft.com/en-us/library/2e08f6yc(VS.71).aspx
【8】Newman?M?E?J.；The?Structure?and?Function?of?Complex?Networks，SIAM?Review，2003，45，Page(s)：167-256
【9】J.A.Pouwelse，et.al；TRIBLER：A?Social-Based?Peer-to-Peer?System；RecentAdvances?in?Peer-to-Peer?Systems?and?Security，2006，Volume?20，Issue?2，Pages：127-138
【10】M.Ripeanu；Peer-to-Peer?Architecture?Case?Study：Gnutella?Network；Proceedingsof?First?International?Conference?on?Peer-to-Peer?Computing，2001，Pages：99-100
【11】Paul?Francis；Is?the?Internet?Going?NUTSS？IEEE?Internet?Computing，November/December?2003，Pages：97-99
【12】Rosenberg?J.，et?al；STUN-Simple?Traversal?of?User?Datagram?Protocol(UDP)Through?Network?Address?Translators(NATs)；RFC?3489，March?2003
【13】Newman?M?E?J.；The?Structure?and?Function?of?Complex?Networks；SIAM?Review，2003，45，Page(s)：167-256
Claims (10)
1. the method for building up based on the Internet event-driven mechanism programming model the steps include:
1) adopt TCP/IP peer protocol pattern to make computing node have the ability of the request of reception and transmission request;
2) between computing node, set up the directly interaction protocol of equity;
3) on computing node, set up message based action listener mechanism, be used to receive data;
4) set up dynamic cluster, and in dynamic cluster, adopt data multicast mechanism to carry out transfer of data; Described dynamic cluster is the computational resource cooperation Share Model that forms in some moment based on the computing node of joint demand;
5) node of monitoring in the described dynamic cluster is transferred to other node with the data of event occurrence node by described data multicast mechanism;
6), set up event-driven programming model according to the data type of transmission in conjunction with software engineering technology.
2. the method for claim 1 is characterized in that described computing node comprises the client and server end.
3. method as claimed in claim 2 is characterized in that the method for building up of the interaction protocol of direct equity between the described computing node is:
1) sets up network and connect monitoring mechanism;
2) pass through the direct connection that IPv6 or NAT crossing technology realize arbitrary node on the Internet;
3), set up the different mutual agreements of client and server end equity according to concrete internet, applications feature.
4. the method for claim 1 is characterized in that utilizing asynchronous programming model to finish the described message based action listener mechanism of setting up in conjunction with ICP/IP protocol.
5. method as claimed in claim 4 is characterized in that the method for work of described message based action listener mechanism is:
After obtaining the related remote event message, start the local event driving mechanism remote events is converted to local incident, drive local related system remote events is responded.
6. method as claimed in claim 5 is characterized in that described related remote incident comprises: message remote events, file remote events, byte remote events and object remote events.
7. the method for claim 1 is characterized in that described data multicast mechanism is that the topological structure that utilizes the BT agreement to form carries out the multicast of lightweight data, and its method is:
1) adopt the state of computing node in the tracker monitoring dynamic cluster of BT agreement, described node state comprises: whether node is in the cluster, node whether to certain incident pay close attention to, whether node be whether super node, node have obtained certain incident;
2) for being in node in the cluster, described tracker equilibrium distributes the multicast task, and its method is: at first for the transmission data unique identification is set; Described then tracker with the super node determined as kind of a child node, by concurrent connection to other node transmission data; The sign of the data that the node inspection receives if repeat, then stops to be transferred to the concurrent connected node of this node; Otherwise the concurrent connected node that is transferred to this node continues multicast.
8. method as claimed in claim 7 is characterized in that sign has the node of process once on the described transmission data, when the node that current arrival node and lightweight data process occur repeats, then stops transmission; Otherwise, continue multicast process.
9. the method for claim 1 is characterized in that described data type comprises: message, file, byte and object.
10. method as claimed in claim 9 is characterized in that described event-driven programming model comprises: message transmission and dependent event, file transmission and dependent event, byte transmission and dependent event and object send and dependent event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810117826A CN101645912B (en) | 2008-08-05 | 2008-08-05 | Internet-based method for building event-driven mechanism programming model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810117826A CN101645912B (en) | 2008-08-05 | 2008-08-05 | Internet-based method for building event-driven mechanism programming model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101645912A true CN101645912A (en) | 2010-02-10 |
| CN101645912B CN101645912B (en) | 2012-10-10 |
Family
ID=41657632
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200810117826A Expired - Fee Related CN101645912B (en) | 2008-08-05 | 2008-08-05 | Internet-based method for building event-driven mechanism programming model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101645912B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103929497A (en) * | 2014-04-30 | 2014-07-16 | 天脉聚源(北京)传媒科技有限公司 | Method for pushing messages in real time and server |
| CN108810030A (en) * | 2018-07-24 | 2018-11-13 | 中国计量大学 | Node property recognition methods based on distributed system |
-
2008
- 2008-08-05 CN CN200810117826A patent/CN101645912B/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103929497A (en) * | 2014-04-30 | 2014-07-16 | 天脉聚源(北京)传媒科技有限公司 | Method for pushing messages in real time and server |
| CN103929497B (en) * | 2014-04-30 | 2017-09-15 | 天脉聚源(北京)传媒科技有限公司 | The method and server of a kind of real time propelling movement message |
| CN108810030A (en) * | 2018-07-24 | 2018-11-13 | 中国计量大学 | Node property recognition methods based on distributed system |
| CN108810030B (en) * | 2018-07-24 | 2020-09-25 | 中国计量大学 | Node property identification method based on distributed system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101645912B (en) | 2012-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111405304B (en) | Anchor interaction method and device, computer equipment and storage medium | |
| US7805407B1 (en) | System and method for dynamic configuration of replicated database servers | |
| CN102571686B (en) | Method for realizing cloud conference system | |
| CN100547973C (en) | A kind of high performance computing system based on peer-to-peer network | |
| Ma et al. | A master-slave blockchain paradigm and application in digital rights management | |
| Zhao et al. | Dynatops: A dynamic topic-based publish/subscribe architecture | |
| Kavalionak et al. | Integrating peer-to-peer and cloud computing for massively multiuser online games | |
| Comi et al. | A qos-aware, trust-based aggregation model for grid federations | |
| Petri et al. | Modelling and implementing social community clouds | |
| CN101098270A (en) | Distributed equity network system | |
| Okoye et al. | Optimized user-friendly transaction time management in the blockchain distributed energy market | |
| CN101645912B (en) | Internet-based method for building event-driven mechanism programming model | |
| Ghosal et al. | P2P contracts: a framework for resource and service exchange | |
| CN101873345A (en) | Integrated block switching P2P shared-file system and method for motivating seed-providing for other nodes thereof | |
| CN101645911B (en) | Dynamic cluster-based news service method | |
| Messeguer et al. | Communication and coordination patterns to support mobile collaboration | |
| CN110061863B (en) | Distributed multi-task crowd-sourcing sensing method based on fairness in sparse network | |
| Stehr et al. | Partially ordered knowledge sharing and fractionated systems in the context of other models for distributed computing | |
| Mesaros et al. | P2PS: Peer-to-peer development platform for mozart | |
| Dhale et al. | Review of cloud computing architecture for social computing | |
| Al-Karkhi et al. | Disruption Recovery within Agent Organisations in Distributed Systems. | |
| CN110061862B (en) | Fairness-based distributed multi-task crowd sensing method in dense network | |
| Carlini | Combining peer-to-peer and cloud computing for large scale on-line games | |
| CN101801111A (en) | Acentric tuple space system facing cooperation | |
| Kakarontzas et al. | Agents, clusters and components: A synergistic approach to the GSP |
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: 20121010 Termination date: 20150805 |
|
| EXPY | Termination of patent right or utility model |