Data flow network. RU patent 2511611.

FIELD: physics, computer technology.

SUBSTANCE: invention relates to computer data processing means. The method involves connecting processes of a program executed in a computer system using a composite model called a source-target structure, which enables to combine operation of the process with data flow networks for forming process networks.

EFFECT: high speed and high throughput of the computer system.

20 cl, 15 dwg

The level of technology

A computer system can provide the ability to run multiple processes at once. The simultaneous execution of processes can provide for the program to run faster or more effective, increasing the capacity of the computer system.

In the processes that implement the model of a control flow logic control flow describes a set of conditions and modifications of the program state and can be arranged, for example, by means of a flowchart sequence of operations. The term "algebra process" is used to refer to the methodology of substantiation of how processes are linked to each other to cooperate in solving the General problem and how managed by the lifetime of the process. Although algebra process may significantly differ from the primitives that modifies the internal state of the process, there is usually a point of interaction, where the communication between processes, provides execution of internal data flow diagrams. Such point of interaction between processes in some programs can be complex. Complex point of interaction may complicate the planning process to run and can lead to less than optimal implementation of the program.

Disclosure of the invention

The section "disclosure of the invention" is intended for input selection concepts in a simplified form, which further described below in the section "Implementation of the invention". The section "disclosure of the invention" is not intended to identify the key characteristics or essential features of the claimed subject matter of the invention and is not intended for use, to limit the scope of the declared object of the invention.

Provided composite model, referred to as the structure of the source-to-target for connection of network processes processes generic, flexible and extensible way. The model provides the possibility of combining overall structural components algebra processes with the networks of the data stream to the establishment of networks of the process. Algebraic operations can be expressed as a composition of a model for formation of networks of a data flow that provide fully interoperable between processes algebraic operations. The model can be applied to a wide array of simple network data flow, as well as to more complex networks of the data stream.

Brief description of drawings

The accompanying drawings included to ensure a more complete understanding of options for implementing and here and be part of this description. The drawings are illustrated options for implementation, and together with the description they serve to explain the principles of embodiments. Other options for implementation and many of the alleged benefits of options for the implementation will be easy to understand as they explanations with reference to the following detailed description of the invention. Elements of the drawings does not necessarily presented in relation to each other. The same numbers of the reference positions marked matching the similar parts.

In Fig. 1A-1C shows the flowchart illustrating ways to implement interfaces, the source and destination, intended for use in network data flow.

In Fig. 2A-2C shows a flowchart illustrating the ways of implementation of the nodes in the network data flow.

In Fig. 3 shows the block diagram illustrating an implementation option network process network data stream that is connected with processes.

In Fig. 4 shows a block diagram that illustrates an implementation option network process to the process that is connected with the networks of the data stream.

In Fig. 5 shows a block diagram that illustrates an implementation option to the originating unit, configured for the implementation of the transfer operation.

In Fig. 6 shows a block diagram that illustrates an implementation option unit trigger configured for the implementation of the receive operation.

In Fig. 7 shows the block diagram illustrating variant of implementation of the network process.

In Fig. 8 shows the block diagram illustrating an implementation option network data stream configured for the implementation of the operation of choice and connections.

In Fig. 9 shows the block diagram illustrating variant of implementation of the network process.

In Fig. 10 shows the block diagram illustrating variant of implementation of the network process.

In Fig. 11 shows the block diagram illustrating the option of implementing a computer system that is configured for implementing made runtime program scheduler process.

The implementation of the invention

In the subsequent detailed description is a reference to the attached drawings, which form part of and are shown to illustrate specific ways of implementation, which can be implemented invention. In this respect the terms representing the direction, such as "top", "bottom", "front", "back", "leading", "normally open", etc. are used with reference to focus on the described drawings (drawings). Because the components of options for implementation may be located in a number of different orientations, terminology is used for the purpose of illustration, but not limitation. You should understand that there may be other options for implementation and structural or logical changes can be made without going beyond the scope of the present invention. Therefore, the following detailed description of the invention should not be viewed as the limit and scope of the present invention applied defined by the claims.

It should be understood that the characteristics of the various approximate methods of implementing described here can be combined with each other, unless specifically noted more.

In Fig. 1A shows the block diagram illustrating the option exercise patterns 10 source-to-target, which includes the interface 12 source and interface 14 goals, United by connecting 16.

Structure 10 source-to-target provides a set of operations that are used to form network data flow (for example, network 32 data flow is shown in Fig. 3 and 4), which connect any number of processes (for example, processes, 40, is shown in Fig. 3 and 4) and a program running on a computer system (for example, a computer system 100, shown in Fig. 11) in the network of the process. Network data flow form the structural components of the algebra of the process, such as "pass", "accept", "select", "connect" and their combinations, to ensure the transfer of data between processes. Structure 10 source-to-target provides a Protocol for transferring data between the interface 12 source and interface 14 goals and provides an embodiment of the structural components of algebra process in the networks of the data stream.

Structure 10 source-to-target provides the connectivity processes when processes have no or minimal information about each other. One process can provide data to another process using the structure 10 source-to-target without any information about another process. Similarly, one process can receive data from another process using the structure 10 source-to-target without any information about another process. Each process includes a set of instructions stored in read into a computer data storage medium (for example, 104 memory, as shown in Fig. 11) and executable by the computer system. Data passed between processes can include any type of data, including pointers, addresses or other structured or encrypted information.

Interface 12 source includes instructions stored on machine-readable media (for example, in the system 104 memory, as shown in Fig. 11) and executable by the computer system for the realization of operations "connect to", "disconnecting from a goal", "reserve", "release" and "eat", where each of operations interface 12 source is called with one or more interfaces 14 the goals. Interface 14 goals includes instructions stored in machine-readable storage medium (for example, 104 memory, as shown in Fig. 11) and executable by the computer system for the realization of operations "sentence", where the operation "proposal" interface 14 goals is invoked with by using one or more interfaces 12 source.

Operation "connection to" interface 12 source establishes a connection between 16 interface 12 source and interface 14 goals, which causes the join order. Operation "detach from the goal of" interface 12 source disconnects the connection between 16 interface 12 source and interface 14 goals, which causes the operation disconnection from the goal. Operation connection to and disconnection from each goal may return indicator of success or failure in the interface 14 goals, which received the call operation, to indicate that it is the connection 16 successfully installed or disconnected.

After a connection is established between 16 interface 12 source and interface 14 goals, interface 12 source causes the operation of the "proposal" interface 14 goals using the connection 16 to supply data in the interface 14 goals. In response to the operation of the "proposal" interface 14 goals returns one of the accepted, rejected, or deferred. Interface 14 goals returns "accepted" to indicate that the interface 14 goals have used the data proposed interface 12 source. Interface 14 goals returns declined to indicate that the interface 14 goals are not interested in the data proposed by the interface 12 source. Interface 14 goals returns "pending" to indicate that the interface 14 goals are not ready to use the data proposed by the interface 12 source, at this point in time.

After the backup of the data interface 14 goals or causes the operation to release the interface 12 source for release early posted by redundancy or operation of consumption for consumption data reserved. Once the interface 12 source confirms data backup interface 14 goals, providing an indicator of successful performance in the interface 14 goals, interface 12 source holds the data for the interface 14 goals, which have placed a reservation, until then, until the data is either released or used interface 14 goals, who placed a reservation.

As shown in Fig. 1B, each interface 14 goals can be connected to any number of interfaces 12(1)-12(M) of the source, causing the join operation with the aim of each interface 12(1)-12(M) to establish a relevant connection 16(1)-16(M), where M is an integer greater than or equal to one and represents the M-th interface 12 source and M-th connection 16. In accordance with this each interface 14 goals can be proposed data from any number of interfaces 12(1)-12(M) of the source.

As shown in Fig. 1C, each interface 12 source can be connected to any number of interfaces 14(1)-14(N) targets in response to the challenges of the join operation with the aim of each interface, 14(1)-14(N) targets, which establish the appropriate connection 16(1)-16(N), where N is an integer greater than or equal to one, and represents the nth interface 14 goals and N-th connection 16. In accordance with this each interface 12 source can provide any number of interfaces 14(1)-14(N) targets.

In Fig. 2A-2C shows a flowchart illustrating ways 20A, 20B 20C and performance respectively site 20 network data flow. As shown in Fig. 2A, site 20A may include one or more interfaces 12(1)-12(P) source, where P is an integer greater than or equal to one, and is P-th interface 12 source, but does not represent either one of the interfaces 14 goals. Sites 20A, with only the interfaces 12 source, form the initiators of the data. Site 20B may include one or more interfaces 14(1)-14(Q) of the target, where Q is an integer greater than or equal to one, and is a Q-t interface 14 goals, but represents none of the interfaces 12 source, as shown in Fig. 2B. Sites 20B, with only the interfaces 14 goals, form data receivers. In addition, the site 20C may include one or more of the interfaces 12(1)-12(P) source and one or more interfaces 14(1)-14(Q) of the target, as shown in Fig. 2C. Each node 20A, 20B 20C and may also include one or more passive processing modules (not shown), made with the possibility of conversion of data as dissemination through the nodes 20A, 20B 20C and.

Data are disseminated via the network data stream by passing from one node to 20 in one or more other nodes 20 using the structure 10 source-to-target. Each connection between two nodes 20 network data stream is presented as the formation of pairs of the site 20 embodying interface 12 source (for example, site 20A or site 20C), and the node that implements the interface 14 goals (for example, site 20B or site 20C).

Any number of network components of the data flow can be identified using the structure 10 source-to-target, and the components can be combined in a largely in an ad hoc network data streams. Examples of networks of data streams include unrelated buffers associated buffers, blocks a single destination buffers producer/consumer, data receivers (a method call), data conversion, timers and processors of input-output. Network data stream can be composed of more complex networks, using the structure 10 source-to-target. For example, the implementation of the release of block read/write or implementation of the monitor can be expressed in respect of networks of a data flow, using the structure 10 source-to-target.

Any number of processes can provide data in network data flow, and any number of processes can receive data from the network data flow, as shown in Fig. 3. In Fig. 3 shows the block diagram illustrating an implementation option network 30 processing with a network of 32 data stream that is connected the processes 40(1)-40(R) and 40(R+1)-40(S), where R is an integer greater than or equal to one, S is an integer greater than or equal to two, R less than S, and is a R-t and S-th process 40.

A network of 32 data flow forms a passive structural element of the data flow, made with the possibility of dissemination of processes 40(1)-40(R) in the processes 40(R +1)-40(S). A network of 32 data flow includes a set of 34 one or more interfaces 14 goals, set 36 of one or more interfaces 12 source and a set of components 38 network, which includes 20 nodes and passive processing units, and mutually connects a set of 34 interfaces 14 goals with a set of 36 interfaces 12 source. Components 38 network can include any appropriate type, number and/or combination of 20 nodes and passive processing units that connect a set of 34 interfaces 14 goals with a set of 36 interfaces 12 source any relevant way.

Processes 40(1)-40(R) incorporate relevant sets 42(1)-42 (R) from one or more interfaces 12 source, where each set of 42 may include the same or a different number of interfaces 12 source, as in other sets 42. A set of 34 interfaces 14 sets goals set 44 connection between 16 34 set of interfaces 14 goals and sets 42(1)-42(R) interfaces 12 source. As shown in Fig. 1B and 1C, each interface 14 goals in set 44 can connect to any number of interfaces 12 source in sets 42(1)-42(R) and any number of interfaces 14 goals in set 44 can connect with each interface 12 source in sets 42(1)-42(R). Each interface 12 source in sets 42 (l)-42(R) provides data dissemination process 40(1)-40(R) in a network of 32 data flow, offering data in one or more interfaces 14 goals in set 44 as described above with reference to Fig. 1A. One or more processes 40(1)-40(R) may also include one or more additional sets of interfaces 12 source (not shown) and/or one or more sets of interfaces 14 goals (not shown), which are connected with one or more other networks 32 data stream (not shown).

Processes 40(R +1)-40 (S) include appropriate sets 46(R+1)-46(S) from one or more interfaces 14 goals, where each set of 46 may include the same or a different number of interfaces 14 goals, as in other sets 46. Sets 46(R+1)-46(S) interfaces 14 goals set of 48 connection between 16 sets 46(R +1)-46(S) interfaces 14 goals and sets from 36 interfaces 12 source. As shown in Fig. 1B and 1C, each interface 14 goals in sets 46(R+1)-46(S) can connect to any number of interfaces 12 source in the set of 36 and any number of interfaces 14 goals in sets 46(R+1)-46(S) can be connected to each interface 12 source in the set of 36. Each interface 12 source in the set of 36 provides for the distribution of data from the network 32 data flow in one or more processes 40(R+1)-40(S), offering data to one or more interfaces 14 goals in sets 46(R+1)-46(S)as described above with reference to Fig. 1A. One or more processes 40(R+1)-40(S) may also include one or more additional sets of interfaces 14 goals (not shown) and/or one or more sets of interfaces 12 source (not shown), which are connected with one or more other networks 32 data stream (not shown).

Processes 40(1)-40(R) can provide data to the processes 40(R+1)-40(S), using a network of 32 data flow without any information about the processes 40(R+1)-40(S). Similarly, processes 40(R+1)-40(S) can take data from processes 40(1)-40(R), using a network of 32 data flow without any information about the processes 40(1)-40(R).

The process can provide data in any number of network data flow and receive data from any number of networks of data streams, as shown in Fig. 4. In Fig. 4 shows a block diagram that illustrates an implementation option network 50 of the process with process 40-connected networks 32(1)-32(T) data flow and 32(T+1)-32(V), where T is an integer greater than or equal to one, V is an integer greater than or equal to two, T is less than V, and is T-th and V-th network of 32 data flow.

The process 40 includes a set of program instructions that are configured to receive data from the network 32(1)-32(T) of the data flow and provide data in network 32(T+1)-32(V) data stream. The process 40 includes a set of 54 from one or more interfaces 14 goals, set 56 one or more interfaces 12 source and a set of components 58 process and mutually connects a set of 54 interfaces 14 goals with a set of 56 interfaces 12 source. Components 58 process may include any of the appropriate type, number and/or combination of instructions and/or data structures that connect a set of 54 interfaces 14 goals with a set of 56 interfaces 12 source any relevant way.

Network 32(1)-32(T) of the data include the appropriate sets 62(1)-62(T) one or more interfaces 12 source, where each set 62 can include the same or a different number of interfaces 12 source, as in other sets 62. A set of 54 interfaces 14 goals establishes a set of 64 connections between 16 54 set of interfaces 14 goals and sets 62(1)-62(T) interfaces 12 source. As shown in Fig. 1B and 1C, each interface 14 goals in 54 set can connect to any number of interfaces 12 source in sets 62(1)-62 (T) and any number of interfaces 14 goals in 54 set can connect with each interface 12 source in sets 62(1)-62(T). Each interface 12 source in sets 62(1)-62(T) provides for the distribution of data from networks 32(1)-32(T) of the data stream in the process 40, offering data in one or more interfaces 14 goals in 54 set as described above with reference to Fig. 1A. Network 32(1)-32(T) of the data each includes one or more sets of interfaces 14 goals (not shown), which are connected with one or more other processes 40 (not shown). Network 32(1)-32 (T) of the data can also include one or more additional sets of interfaces 12 (not shown) and/or one or more sets of interfaces 14 goals (not shown), which are connected with one or more other processes 40 (not shown).

Network 32 (T+1)-32 (V) data flow include the appropriate sets 66(T+1)-46(F) one or more interfaces 14 goals, where each set 66 can include the same or a different number of interfaces 14 goals that other sets 66. Sets 66(T+1)-46(V) interfaces 14 goals set 68 connections between 16 sets 66(T+1)-46(V) interfaces 14 goals and a set of 56 interfaces 12 source. As shown in Fig. 1B and 1C, each interface 14 goals in sets 66(T+1)-46(V) can connect to any number of interfaces 12 source in the set of 56, and any number of interfaces 14 goals in sets 66(T+1)-46(V) can connect with each interface 12 source in the set of 56. The process 40 provides for the distribution of data in one or more networks 32(T+1)-32(V) data flow, providing each interface 12 source in the set of 56 offer data in one or more interfaces 14 goals in sets 66(T+1)-46 (V)as described above with reference to Fig. 1 A. Network 32(T+1)-32(V) data stream can also include one or more sets of interfaces 12 source (not shown) and/or one or more sets of interfaces 14 goals (not shown), which are connected with one or more other processes 40 (not shown).

In Fig. 5 shows a block diagram that illustrates an implementation option block 70 initiator, made with the possibility of realization of transfer operations. Block 70 initiator includes interface 12 source, but does not include any of the interfaces 14 goals. Block 70 initiator is included in the process and 40 provides for the distribution in the process 40 data in a network of 32 data flow. In accordance with this block 70 initiator embodies the transfer operation by receiving data from the process 40 and provide the data in one or more connected networks 32 data flow.

Block 70 initiator can interact or may not interact with the main process scheduler that schedules processes 40 to run in a computer system. Thus, the block 70 initiator may not include an interface to the scheduler.

In Fig. 6 shows a block diagram that illustrates an implementation option block 80 trigger, made with the possibility of realization of the receive operation. The trigger 80 includes interface 14 goals, but does not include any of the interfaces 12 source. Unit 80 trigger included in the process and 40 provides the ability to receive data in the process 40 from the network 32 data flow. Unit 80 trigger acts as the endpoint end or one of any number of endpoints for networks 32 data flow. In accordance with this unit 80 trigger embodies the operation of reception by receiving data from the network 32 data flow and provide these data to process 40.

Unit 80 trigger also includes interface 82 scheduler. Interface 82 scheduler interacts with the main process scheduler that schedules the process 40 to run in a computer system. In particular, the interface 82 of the scheduler is the task scheduler, so you can install a pause in the process 40 (e.g., blocking) in response to receiving the entries containing block start 80 trigger when data is not available immediately. Interface 82 scheduler also works with the scheduler, allowing the resumption of the process 40 (for example, unlocking) in response to these operations reception, which includes the achievement of the block 80 trigger block 80 trigger.

In Fig. 7 shows the block diagram illustrating an implementation option network 90 processing, which includes operations "send" and "receive". In Fig. 7 the process 40(1) receives the data from the process 40(2) and provides data in the process 40(3). To receive data from the process 40(2) block 80(1) of the trigger process 40(1) provide data from a block 70(1) of the initiator of process 40(2) and block 80(1) trigger receives the data, as indicated by the network 32A data stream, which embodies the receive operation. The process 40(1) performs the calculation in respect of the data and sends the resulting data in the process 40(3). Data in the process 40(3) block 70(2) the initiator of the process 40(1) provides data in a block of 80(2) of the trigger process 40(3), and block 80(2) trigger receives the data, as indicated by 32B network data flow, which embodies the transfer operation.

In Fig. 8 shows the block diagram illustrating an implementation option networks 32 data flow, made with the possibility of realization of selection and connections. A network of 32 data flow includes sites 20(1)-20(W+1), where W is an integer greater than or equal to two, and W is the first node 20, W-th process and 40 W block 70 initiator. Each node 20(1)-20(W+1) includes interface 12 source and interface 14 goals. Interfaces 14 goals in knots 20(1)-20(W) taking data from the corresponding blocks 70(1)-70(W) of the initiator in the relevant processes 40(1)-40(W). Interfaces 12 source in knots 20(1)-20(W) each provides the data interface 14 goals in the site 20(W+1). Interface 12 source node 20(W+1) provides data in block 70 trigger during 40(W+1).

For realization of operations selecting a site 20(W+1) only accepts data from the first node 20(1)-20(W), c in order to offer the data in the node 20(W+1), and rejects the proposals of the data from all the other nodes 20(1)-20(W). Site 20(W+1) provides data in a block of 80 trigger during 40(W+1).

In one embodiment, each node 20(1)-20(W) made use of its appropriate interface 12 source for the quotation of its identity as the data in the node 20(W+1) in response to the proposal of data from the relevant units 70(1)-70(W) initiator. Site 20(W+1) uses its interface 14 goals in order to take the first identity proposed set of nodes 20(1)-20(W), and to reject all future of identity, the proposed set of nodes 20(1)-20(W). Site 20(1)-20(W), which accepts an answer on the acceptance of the site 20(W+1), using its interface 12 source, takes the data proposed from the corresponding block 70(1)-70(W) of the initiator using your interface 14 goals. Nodes 20(1)-20(W), who are the answers-deviations from the site 20(W+1), using their respective interfaces 12 source, reject the data proposed from the corresponding block 70(1)-70(W) of the initiator using their respective interfaces 14 goals. Site 20(W+1) uses its interface 12 source in order to offer adopted identity as the data in block 70 trigger during 40(W+1). Block 70 trigger responds by accepting the identity of the site 20(W+1). The process 40(W+1) then retrieves the data selection from the site 20(1)-20(W)using identity, adopted from the site 20(W+1).

In another embodiment, the can be excluded hosts 20(l)-20(W) and site 20(W+1) can accept data directly from blocks 70(1)-70(W) of the initiator. In this embodiment, the blocks 70(1)-70(W) of the initiator of the offer the data in the node 20(W+1). Site 20(W+1) uses its interface 14 goals in order to take the first data suggested from the set of blocks 70(1)-70(W) initiator, and to reject all subsequent data suggested from the set of blocks 70(1)-70(W) of the initiator. Site 20(W+1) uses its interface 12 source for proposals received data in block 70 trigger during 40(W+1). Block 70 trigger is responsible, taking data from node 20(W+1) to complete the operation of choice.

In other versions of the operation of choice can be embodied in the network 32C of the data flow, using other combinations of patterns 10 source-to-target, which is the interface with other combinations of blocks 70 initiator and/or 80 blocks the trigger.

In Fig. 9 shows the block diagram that illustrates an implementation option network 92 process, which includes the operation of choice. In Fig. 9 selecting operation embodied between the process 40(4) and processes 40(5) 40(6), by incorporating network 32C of the data stream. Using network 32C data flow process 40(4) accepts data from the first of processes 40(5) 40(6), which provide data in network 32C of the data stream. The process 40(4) performs some calculations on the data and sends the resulting data in the process 40(7)using the network 32D data stream, which embodies the transfer operation.

An implementation option, shown in Fig. 8, can also be used for the realization of join operations. For realization of operations connections node 20(W+1) waits to receive data from all nodes 20(1)-20(W) before you will be offered the data from all nodes 20(1)-20(W) in the block 80 trigger during 40(W+1).

In one embodiment, each node 20(1)-20(W) are executed with the possibility of offering their identity as part of the data, the proposed site 20(W+1), using the interface of 12 source in response to the proposal of data from the relevant units 70(1)-70(W) initiators. Site 20(W+1) keeps track of the proposals adopted, using your interface 14 goals, to determine, when all nodes 20(1)-20(W) offered their identity. Site 20(W+1) uses its interface 14 goals to answer all proposals, with the exception of the last sentence, which was delayed. In response to suggestions from all adopted nodes 20(1)-20(W), site 20(W+1) uses its interface 14 goals to call a backup operation on all nodes 20(1)-20(W), with the exception of the site 20(1)-20(W), which provided the last sentence. Site 20(W+1) waiting for a response in the host 20(1)-20(W), which provided the latter proposal, adopted until the site 20(W+1) determines whether the reservation was successful, or any of the reservations were unsuccessful.

All nodes 20(1)-20(W), who took the call for backup, use their respective interfaces 14 goals to call backing for the relevant blocks 70(1)-70(W) of the initiator. Blocks 70(1)-70(W) initiator answer calls reservation, providing the indicator is either successful or unsuccessful execution to the appropriate call nodes 20(1)-20(W). If the block 70 initiator no longer contains the proposed data or proposed data have been reserved by another host 20 (not shown), block 70 initiator answers the call reservation indicator of failure. Otherwise block 70 initiator answers the call reservation indicator of successful implementation.

Site 20(W+1) detects any failure reservation in response to the reception of indicators for the failure of the nodes 20(1)-20(W) or in response to disable any of the nodes 20(1)-20(W). If the node 20(W+1) detects any failure redundancy, site 20(W+1) uses its interface 14 goals to invoke the operation of liberation for all nodes 20(1)-20(W), who answered the call of reserving success indicators. Site 20(W+1) returns the index of pending execution site 20(1)-20(W), who presented the latest proposal, as adopted by the site 20(W+1). In response to receiving a call release sites 20(1)-20(W) release successfully made the reservation and, in turn, cause the operation of the release to the blocks 70(1)-70(W) of the initiator. Blocks 10(1)-70(W) initiator release successful reservation in response to receiving a call for the release of 20 knots(1)-20(W).

If the node 20(W+1) detects that all reservations are completed successfully, the site 20(W+1) uses its interface 14 goals to invoke the operation of consumption for all nodes 20(1)-20(W), with the exception of the site 20(1)-20(W), which provided the latter proposal, as adopted by the site 20(W+1). Site 20(W+1) returns "accepted" in the site 20(1)-20(W), which provided the last sentence, and this site 20(1)-20(W), in turn, returns the "accepted" in the corresponding block 70(1)-70(W) of the initiator. All nodes 20(1)-20(W), which accepted the challenge on consumption, in turn, use their respective interfaces 14 goals to invoke the operation of consumption for the relevant blocks 70(1)-70(W) of the initiator. Site 20(W+1) uses its interface 12 source in order to offer combined in block 70 trigger during 40(W +1). Block 70 trigger responds by taking data from a site 20(W+1) to complete the joint operation.

In another embodiment, the site 20(W +1) sends signals to block 70 trigger during 40(W+1), using the Boolean value, when all the data has been reserved or accepted in knots 20(1)-20(W). Block 70 trigger receives data from nodes 20(1)-20(W). In this embodiment, the site 20(W+1) returns a legend in the last node 20(1)-20(W)that provide data.

In other variants of implementation of the joint operation can be incorporated in a network 32C of the data flow, using other combinations of patterns 10 source-to-target, which is the interface with other combinations of blocks 70 initiator and/or 80 blocks the trigger.

In Fig. 10 shows the block diagram illustrating an implementation option network 94 process, which includes a join operation. In Fig. 10 operation connection embodied between the process 40(8) and processes 40(9) and 40(10) by including them in the network 32C of the data stream. Using network 32C data flow process 40(4) accepts data provided from each of the processes 40(9) and 40(10) in the network 32C of the data stream. The process 40(8) performs some calculations using the data, and sends the resulting data in the process 40(11), using the network 32E data stream, which embodies the transfer operation.

Operations such as "transfer", "acceptance", "choice" and "connection", presented above with reference to figure 5-10, can be combined into more complex configuration, in other variants of execution.

In Fig. 11 shows the block diagram illustrating the option of implementing a computer system is 100, which is made with the possibility of realization of Executive software environment, which includes the scheduler process.

Computer system 100 includes one or more packages 102 processor, system 104 memory, zero or more devices 106 I/o, zero or more devices 108 display, zero or more peripheral devices 110 and zero or more network devices 112. Packages 102 processor, the system 104 memory device 106 I/o device 108 display, peripherals 110 and network devices 112 linked, using a set of mutual connections 114, which include any appropriate type, the number and configuration of controllers, tires, interfaces, and/or other wired or wireless connections.

A computer system is 100 any relevant processing device, made with the possibility of its use for General purpose or special purpose. Examples of computer systems 100 include the server, personal computer, portable computer, tablet PC, personal digital assistants (PDA, pocket PC, mobile phone and audio/video device. Components of a computer system 100 (that is, packets 102 processor, the system 104 memory devices 106 I/o device 108 display, peripherals 110, network devices 112 and mutual connection 114) may be contained in the General case (not shown) or in any appropriate number of individual buildings (not shown).

Computer system 100 loads and executes OS 120. OS 120 includes statements that are executed hardware threads 116 for managing the components of a computer system 100, and provides a set of functions that provide access to and use applications 124 components. In one embodiment, the OS 120 represents the Windows operating system. In other versions of the OS 120 represents another operating system suitable for use with the computer system 100.

The level 121 resource management includes instructions that can be executed in conjunction with OS 120 to allocate the resources of a computer system 100, including hardware threads 116. The level 121 resource management can be included in a computer system 100 as library functions available for one or more applications 124, or as an integrated part of the OS 120.

Executive platform 122 includes instructions that can be executed in conjunction with OS 120, and the level 121 resource management for generating Executive software environment, and providing Executive functions in applications 124. These Executive functions include the schedule function. When called, the function scheduler generates scheduler that during the work planning processes programs, such as applications 124, to perform one or more hardware threads 116(1)-116(A). Executive functions can be included in a computer system 100, as part of the application 124 as library functions available for one or more applications 124, or as an integral part of the OS 120 and/or level 121 resource management.

Each application 124 includes instructions that can be executed in conjunction with OS 120, level 121 resource management and/or Executive platform 122, to ensure the required operations computer system 100. Each application 124 represents one or more programs that can be executed in the scheduler provided by the Executive platform 122.

The system 104 memory includes any relevant type, quantity and configuration volatile or non-volatile external drives, made with the ability to store instructions and data. Device-drive system 104 memory present a machine-readable media that retain CPU instructions, including OS 120, level 121 resource management, Executive platform 122 and 124 application. Instructions can be executed by a computer system to perform functions and methods OS 120, level 121 resource management, Executive platform 122 and applications 124 described here. Examples of devices-hard drives in the system 104 memory include hard disk drives, random access memory (RAM), permanent memory (ROM), drives and card storage devices like flash and magnetic and optical disks.

The system 104 memory stores instructions and data received from package 102 CPU, devices, 106 input/output devices 108 display, peripherals 110 and network devices 112. The system 104 memory provides saved instructions and data into packets 102 processor, device 106 I/o device 108 display, peripherals 110 and network devices 112.

Device 106 I/o include any appropriate type, number and configuration device I/o, made with the possibility of entering instructions or data from a user's computer system 100 and withdrawal instructions or data from a computer system 100 user. Examples of devices 106 I/o include a keyboard, mouse, touch panel, touch screen, buttons, decorative disks, keys and switches.

Device 108 display can include any of the appropriate type, the number and configuration of display devices made with the possibility of the output of text and/or graphic information to the user of a computer system 100. Examples of devices 108 display include the monitor, the display screen and the projector.

Peripherals 110 include any appropriate type, the number and configuration of peripherals designed to work with one or more other components in a computer system 100, for General or specific processing functions.

Network device 112 include any appropriate type, the number and configuration of network devices, made with the possibility of software for computer systems 100 data transfer through one or more networks (not shown). Network device 112 can work in accordance with any appropriate network Protocol and/or configuration to enable transfer of information computer system 100 in the network or the receiving computer system 100 information from the network.

In an embodiment shown in Fig. 11 one or more of the OS 120, level 121 resource management, Executive platform 122 and applications 124 may include network process with the processes 40, connected with the use of networks 32 data flow, which include any appropriate number of structures 10 source-to-target, as described above with reference to figures 1A-10.

The above options for implementation provide a composite model, called the structure of the source-to-target, for connection of network processes processes, in General, flexible and extensible way. The model allows to combine General constructs algebra process network data flow to the establishment of networks of the process. Typical algebraic operations, such as send, receive, select the connection that can be expressed on the basis of a composite model for the formation of networks of a data flow that provide a fully executed with the possibility of interaction between processes algebraic operations. This model can be applied in a wide array of simple network data flow, as well as in more complex networks of the data stream.

The model also allows the use of loosely connected network of the process. Using loosely connected network of the process, the process may not have or may have limited information on other processes with which this process interacts. Instead, the process has information on connections with other processes, and importance of these connection points.

Loosely connected network of the process can be used to achieve highly suitable for service programs and allow the use of more complex and sophisticated routing logic-based constructs a data stream. Passive network data flow can more effectively to send data from one process to another, than the active processes, acting as intermediaries.

Although the specific ways of implementation were presented and described here is for specialists in the art it will be clear that the different alternative and/or equivalent options for implementation may be presented to replace specific shows and describes options for the implementation, without going beyond the scope of the present invention. This application is designed for coverage of any adaptations or variations specific options for the implementation described here. Therefore, it is assumed that this invention is limited only by the formula of the invention and its equivalents.

1. Way to connect the processes of a programme in a computer system using a set of operations provided by the structure of the data source-to-target, and the structure of the data source-to-target contains interface data source and a data interface objective, whereas the mentioned method contains a call first join operation interfaces in the first interface data source with first data interface goal, which forms the first connection interfaces between the first activity data source and the first data interface objectives; the call of the first operation offers data in the first interface data transfer targets for proposals the first data from the first interface data source in the first data interface objectives; and a call first backing up the data in the first interface data source with the first interface data transfer target after calling the first operation offers data.

2. The method according to claim 1, further comprising: return "pending" from the first data interface goal in the first interface data source in response to the first proposal of the data; and the implementation call first backing up the data in the first interface data source with first data interface targets for booking of the first data first data interface goal after returning "pending".

3. The method of claim 2, additionally contains a call operations consumption data in the first interface data source with first data interface targets for the consumption of the first data first data interface goal after calling the first backing up data.

4. The method of claim 2, additionally contains a call operation of the release of the data in the first interface data source with first data interface goal for the release of the first data after calling the first backing up data.

5. The method according to claim 1, further contains a call to the second connection interfaces in the second interface data source with first data interface goal, which forms a second connection interfaces between the second operation data source and the first data interface objectives; call the second operation offers data in the first interface data transfer targets for proposals second data from the second interface data source in the first data interface objectives; and call the second backing up data in the second transmission interface data source with first data interface goal after a call to the second surgery offers data.

6. The method according to claim 5, additionally contains a call first operation consumption data in the first interface data source with first data interface targets for the consumption of the first data first data interface goal after calling the first backing up data; and call the second operation consumption data in the second interface data source with first data interface targets for the consumption of the second data first data interface after calling the second backing up data.

7. The method according to claim 5, additionally contains a call first operation of the release of the data in the first interface data source with first data interface goal for the release of the first data after calling the first backing up data; and call the second the operation of the release of the data in the second interface data source with first data interface goal for the release of the second data after a call to the second backing up data.

8. Machine-readable medium of storage for connection of the processes of a programme in a computer system using a set of operations provided by the structure of the data source-to-target, and the structure of the data source-to-target contains interface data source and a data interface goal containing executable computer instructions that when the performance of their computer system perform the way, containing: provision of first data from the first software process network data flow using the first structure of the data source-to-target, which includes, at least, join operation interfaces, suggestions data, data backup, release data and consumption data; provision of first data from the network data flow in the second software process using the second structure data source-to-target, which includes, at least, join operation interfaces proposals of data, data backup, release data and consumption data;

9. Machine-readable medium of storage of claim 8, and mentioned method additionally includes: provision of data from the second third of the programming process in network data flow using a third structure data source-to-target, which includes, at least, join operation interfaces, suggestions, data, data backup, release data and consumption data; and provide the first data from the network data flow in the second software process using the second the structure of the data source-to-target in response to the provision of the first data into network data stream before the second data will be provided in the network data flow.

10. Machine-readable medium of storage of claim 8, and mentioned method additionally includes: provision of data from the second the third software process network data flow using a third structure of the data source-to-target, which includes, at least, join operation interfaces, suggestions, data, data backup, release data and consumption data; and providing first data and data from the second network data flow in the second software process using the second structure data source-to-target after submission of data to first and second data into network data flow.

11. Machine-readable medium of storage of claim 8, and a way of additionally contains the first conversion of data into network data flow before providing the first data from the network data flow in the second software process.

12. Machine-readable medium of storage of claim 8, and mentioned method contains the implementation pause in the implementation of the second programme process before providing the first data from the network data flow in the second software process.

13. Machine-readable medium of storage of claim 8, and the mentioned method also includes the provision of first data from the network data flow in the third software process using the third structure of the data source-to-target, which includes, at least, join operation interfaces, offers data backup data release data and consumption data.

15. The method according to 14, additionally contain: the establishment of the third connection interfaces between the third interface data source in the third software process, which includes, at least, backing up data release and data consumption data and first data interface goal in the network data flow.

16. The method according to 14, additionally contain: the establishment of the third connection interfaces between the third interface data source software in the first process, which includes at least, backing up data, release data and consumption data, and the third data interface goal in the network flow data, which includes at least the operation offers data.

17. The method according to 14, additionally contain: the establishment of the third connection interfaces between the third interface data source in the third software process, which includes, at least, backing up data, release data and consumption data, and the third data interface goal in the network data flow, including least, the operation offers data.

18. The method according to 14, additionally contain: the establishment of the third connection interfaces between the third interface data source on the network flow data, which includes, at least, backing up data, release data and consumption data, and the second data interface goal in the second software process.

19. The method according to 14, additionally contain: the establishment of the third connection interfaces between the third interface data source on the network flow data, which includes at least, backing up data, release data and consumption data, and the third data interface goal in the second software process, which includes at least the operation offers data.

20. The method according to 14, additionally contains: the establishment of the third connection interfaces between the third interface data source on the network flow data, which includes, at least, backing up data, release data and consumption data, and the third data interface goal in the third software the process, which includes at least the operation offers data.