CN104243579A - Computational node control method and system applied to water conservancy construction site - Google Patents

Abstract

The present invention proposes a method for controlling computing nodes applied to water conservancy construction sites, including the following steps: using periodic polling to find multiple computing nodes that can be used for computing tasks; obtaining the current computing capabilities of multiple computing nodes respectively, and The computing tasks are decomposed, and the decomposed computing tasks are collaboratively processed by multiple computing nodes; each computing node sends the processing results to the central control node; the central control node analyzes the processing results of each computing node to process the multiple computing nodes control. The method of the invention makes full use of the remaining computing power of each computing node (such as a sensor and a data processing unit) on the construction site, and can effectively improve the informatization level of the water conservancy construction site. The invention also provides a control system for computing nodes applied to water conservancy construction sites.

Description

Control method and system for computing nodes applied to water conservancy construction sites

technical field

The invention relates to the technical field of distributed computing, in particular to a control method and system for computing nodes applied to water conservancy construction sites.

Background technique

With the rapid popularization of the Internet of Things and sensor networks, the use of sensor networks in construction sites is increasing. These sensor networks are widely used to collect various business-related aspects such as temperature, humidity, pressure, and personnel location information, and with the development of management to digitization and informationization, it has also laid a solid foundation for the introduction and development of other businesses . However, for a long time, each network and its nodes have performed their own duties and are separated from each other, unable to achieve the purpose of pervasive computing and information fusion. For example, some CPUs use 32-bit modern CPUs, but the occupancy rate has been below 1% for a long time , so that the computing potential is far from being brought into play, and the central server is overloaded when encountering large computing tasks and computing-intensive operations, and the computing cycle is too long, which affects real-time efficiency decision-making.

However, there are very few solutions to the above problems at present, some of which only mention some simple designs, and some of them only propose a little idea, and there is no real solution that can be used to develop a distributed computing.

Contents of the invention

The present invention aims at solving one of the technical problems in the related art mentioned above at least to a certain extent.

For this reason, an object of the present invention is to propose a kind of control method that is applied to the calculation node of water conservancy construction site, and this method has fully utilized the remaining computing power of each calculation node (as sensor and data processing unit) of construction site, can effectively Improve the informatization level of water conservancy construction sites.

Another object of the present invention is to provide a control system for computing nodes applied to water conservancy construction sites.

In order to achieve the above object, the embodiment of the first aspect of the present invention proposes a control method for computing nodes applied to water conservancy construction sites, including the following steps: using periodic polling to find multiple computing nodes that can be used for computing tasks; respectively Obtaining the current computing capabilities of the multiple computing nodes, decomposing the computing tasks, and co-processing the decomposed computing tasks through the multiple computing nodes; each computing node sends the processing results to the central control node; The central control node analyzes the processing results of each computing node to control the multiple computing nodes.

In addition, the method for controlling computing nodes applied to water conservancy construction sites according to the above-mentioned embodiments of the present invention may also have the following additional technical features:

In some examples, the use of periodic polling to discover multiple computing nodes that can be used for computing tasks specifically includes: sending a polling request according to the computing node list and enabling a waiting timer; each computing node receives the polling request, Estimate their current computing power and send it to the central control node, including:

M=N+P1+P2,

Wherein, M is the current computing capability of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate of the past period of time, and P2 is the CPU occupancy rate of the expected future period of time; before the waiting timer expires, all The central control node judges whether multiple computing nodes can complete the computing task according to the current computing capabilities of each computing node; if yes, use the multiple nodes to complete the computing task, otherwise continue to send polling requests; when When the waiting timer expires, no longer wait for a response from the computing node, and discard the timed-out response message.

In some examples, the obtaining the current computing capabilities of the multiple computing nodes respectively, decomposing the computing tasks, and co-processing the decomposed computing tasks through the multiple computing nodes specifically includes: setting the A plurality of computing nodes is N, and decomposing the computing task into m subtasks, wherein N>M; sending each subtask to a corresponding computing node, and starting an overtime timer; regularly judging whether each computing node fails; Before the timeout timer expires, the calculation results of each calculation node are received.

In some examples, it also includes: adopting a redundant strategy, the same decomposed subtask can be assigned to multiple computing nodes.

In some examples, the communication protocol in XML format is adopted between the computing nodes.

According to the control method of computing nodes applied to water conservancy construction sites according to the embodiments of the present invention, the central control node initiates regular polling, and the potential participating nodes report their respective remaining computing capabilities, and perform task decomposition according to the data reported by each node, Assign to designated nodes for calculation, and report the calculation results, and finally summarize the final results based on the information reported by each node. Therefore, this method makes full use of the remaining computing power of each computing node (such as sensors and data processing units) on the site, and can effectively improve the informatization level of the water conservancy construction site.

The embodiment of the second aspect of the present invention provides a computing node control system applied to a water conservancy construction site, including: a discovery module, the discovery module is used to discover multiple computing nodes that can be used for computing tasks through periodic polling an allocation module, the allocation module is used to respectively obtain the current computing capabilities of the multiple computing nodes, and decompose the computing tasks, and cooperatively process the decomposed computing tasks through the multiple computing nodes; the reporting module, The reporting module is used to transmit the processing result of each computing node; the control module analyzes the processing result of each computing node to control the multiple computing nodes.

In addition, the control system applied to the calculation node of the water conservancy construction site according to the above-mentioned embodiments of the present invention may also have the following additional technical features:

In some examples, the discovery module discovers multiple computing nodes that can be used for computing tasks through periodic polling, specifically including: sending a polling request according to the computing node list and enabling a waiting timer; each computing node receives the polling request, estimate their current computing power, and send it to the control module, including:

M=N+P1+P2,

Wherein, M is the current computing capability of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate of the past period of time, and P2 is the CPU occupancy rate of the expected future period of time; before the waiting timer expires, all The control module judges whether multiple computing nodes can complete the computing task according to the current computing capabilities of each computing node; if yes, use the multiple computing nodes to complete the computing task, otherwise continue to send polling requests; when all When the waiting timer expires, the discovery module no longer waits for the response from the computing node, and discards the timed-out response message.

In some examples, the allocating module respectively acquires the current computing capabilities of the multiple computing nodes, decomposes the computing tasks, and cooperatively processes the decomposed computing tasks through the multiple computing nodes, specifically including: setting The plurality of computing nodes is N, and decomposing the computing task into m subtasks, wherein N>M; sending each subtask to a corresponding computing node, and starting an overtime timer; regularly judging whether each computing node Invalidation: before the timeout timer expires, the calculation results of each calculation node are received.

In some examples, the allocation module is further configured to adopt a redundant strategy, and the same decomposed subtask can be allocated to multiple computing nodes.

In some examples, the communication protocol in XML format is adopted between the computing nodes.

According to the control system of the computing nodes applied to the water conservancy construction site according to the embodiment of the present invention, the central control node initiates regular polling, and the potential participating nodes report their respective remaining computing capabilities, and perform task decomposition according to the data reported by each node, Assign to designated nodes for calculation, and report the calculation results, and finally summarize the final results based on the information reported by each node. Therefore, the system makes full use of the remaining computing power of each computing node (such as sensors and data processing units) on site, which can effectively improve the informatization level of water conservancy construction sites.

Additional aspects and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and comprehensible from the description of the embodiments in conjunction with the following drawings, wherein:

Fig. 1 is a flow chart of a control method applied to a computing node at a water conservancy construction site according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of four stages of a control method applied to a computing node at a water conservancy construction site according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a discovery phase according to an embodiment of the present invention;

4 is a schematic diagram of a maintenance information model of a central control node according to an embodiment of the present invention; and

Fig. 5 is a structural block diagram of a control system applied to computing nodes at a water conservancy construction site according to an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

A control method and system applied to computing nodes at water conservancy construction sites according to embodiments of the present invention will be described below with reference to the accompanying drawings.

Fig. 1 is a flow chart of a control method applied to a computing node at a water conservancy construction site according to an embodiment of the present invention. As shown in Figure 1, the control method applied to the calculation node of the water conservancy construction site according to an embodiment of the present invention includes the following steps:

Step S101, using periodic polling to discover multiple computing nodes that can be used for computing tasks.

Specifically, in some examples, as shown in Figure 3, this step specifically includes:

Step 1: Send a polling request based on the compute node list and enable the waiting timer. In other words, the central control node sends polling requests one by one according to the list of Most Recently Used (MRU) nodes in the computing node table maintained by the last calculation, and waits for the computing node (recorded as Nn) to respond, and at the same time Turn on the waiting timer, denoted as Tn.

In some examples, preferably, the node table is in the form of a linked list, and its format is as follows:

Linked list head, total number-→node ID, node addressing→next node...→linked list tail.

In addition, in some examples, the central control node maintains an information table, which records the situation of each computing node, including the correspondence between the logical number and the physical number. Correspondingly, for the computing node whose physical number fails, the Delete accordingly in the information table. In other words, the node information includes the correspondence between the logical number and the physical number. The logical number refers to the ID of the computing node, and the physical number refers to the actual representation of the computing node. For example, for the network card, it is the Mac number. Correspondingly, for the physical Compute nodes with invalid serial numbers are also deleted from the information table. In a specific example, the general message format is expressed in a table as shown in Table 1 below:

serial number marked illustrate 1. message number message type 2. source node number node number 3. destination node number destination node number 4. version type The version number of this message type 5. timestamp Stores the number of seconds elapsed since 12:00:00 on January 1, 2000

6. serial number An increasing serial number 7. message body The message content corresponding to the message type 8. check code Algorithm using validation

Table 1

The actual content of a message is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<version>1.0</version>

<src_node>1</src_node>

<dest_node>2</dest_node>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>no</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Further, the central control node maintains more information, such as the information model diagram shown in Figure 4, which includes the information content shown in Table 2 below:

Table 2

In some examples, preferably, the polling request sent is:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<src_node>1</src_node>

<dest_node>2</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>no</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

In this step, the waiting timer may be set to 10s, for example.

Step 2: Each computing node receives the polling request, estimates its current computing power, and sends it to the central control node, including:

M=N+P1+P2,

Among them, M represents the current computing capability of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate in the past period of time, and P2 is the expected CPU occupancy rate in the future period of time.

Further, in some examples, the format of the sent response message is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_ACK</message_id>;

<src_node>2</src_node>

<dest_node>1</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>1</seq_no>

</head>

<body>

<power>234</power>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Step 3: Before the waiting timer expires, the central control node judges whether multiple computing nodes can complete computing tasks according to the current computing capabilities of each computing node.

Step 4: If possible, use multiple nodes to complete the computing task, otherwise continue to send polling requests. That is, the central control node judges whether these computing nodes participating in this computing task (in consideration of redundancy) can complete the computing task according to the situation reported by each computing node. If it can be completed, it enters the assignment stage, that is, executes step S103 . If it cannot be completed, the polling is continued to enter a larger-scale discovery phase, and the point-to-point message in step 1 is changed to a broadcast polling request message. In some examples, the specific message format is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<src_node>1</src_node>

<dest_node>2</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>yes</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Step 5: When the waiting timer expires, no longer wait for the response from the computing node, and discard the timed-out response message.

In step S102, the current computing capabilities of multiple computing nodes are obtained respectively, and the computing tasks are decomposed, and the decomposed computing tasks are processed cooperatively by the multiple computing nodes.

Specifically, this step specifically includes:

Step A: Set the number of computing nodes as N, and decompose the computing task into m subtasks, where N>M. For example, multiple computing nodes are respectively N1, N2, ... Nn, and the computing task T is decomposed into m subtasks respectively T1, T2, ... Tm, and N>M, then the computing nodes N1, N2, ... Nn respectively process the subtasks correspondingly. Tasks T1, T2, ... Tm.

In some examples, the expression of the subtasks in this step is semantic, so that in a heterogeneous environment, it has nothing to do with the operating system where the computing nodes are located, and the above decomposed subtasks are uniformly described by MathML.

Step B: Send each subtask to the corresponding computing node, and start the timeout timer.

Step C: Maintain the packet liveness (heartbeat) timer H of each computing node, and regularly determine whether each computing node is invalid.

Step D: Receive the calculation results of each calculation node before the timeout timer expires.

It should be noted that, in the above process, for the purpose of fault tolerance, the target computing node simply discards the packets that do not match its own ID.

Step S103, each computing node sends the processing result to the central control node.

In step S104, the central control node analyzes the processing results of each computing node to control multiple computing nodes.

In summary, the method of the present invention makes full use of the remaining computing power of a large number of various sensors and data processing units (that is, computing nodes) laid on the site, and forms a network through a suitable protocol, and the protocol can be mainly summarized as four The phases are: discovery, assignment, reporting and summarization, as shown in Figure 2.

Specifically, in the discovery phase, the central control node initiates regular polling to discover computing nodes that can be used for this computing task; in the assignment phase (allocation phase), the computing The tasks are decomposed and calculated at the same time; the reporting stage is that the computing nodes assigned subtasks report the calculation results to the central control node through the network. Among them, there are two kinds of reported results, one is calculation success, the other is failure, of course Since it is carried out through the network, the central control node needs to detect the failed computing node. If the assigned subtask cannot be completed within the specified time, the computing node will not report; the summary stage is to analyze and summarize the above reported results. For The failed results are sent to another valid computing node while waiting for the calculation results. The specific steps are similar to the allocation stage, which belongs to the secondary allocation. In even more extreme cases, the second assignment will not yield good results, and you can try multiple times until the set time is reached or the number of attempts is reached, and the calculation task is invalidated.

In some examples, in consideration of node failure, the method of the present invention adopts a redundancy strategy, and the same decomposed subtask can be allocated to multiple computer nodes. This way, it is also possible to compare the calculation results of compute nodes performing the same task.

In one embodiment of the present invention, the communication protocol between computing nodes adopts XML (Extensible Markup Language, Extensible Markup Language) format and is standardized within large enterprises or follows corresponding international national standards. Specifically, as far as the prior art is concerned, in the large-scale multi-unit distributed system of an enterprise, the usual method for data exchange between application units is generally through the network socket (Socket) protocol, and basically a custom report is adopted. However, due to the lack of uniform standards, these self-defined information formats are arbitrary, versatile, and inflexible, and cannot meet the reality of long IT construction cycles and emerging new technologies in enterprises. demand. Therefore, the present invention adopts a standard communication protocol in XML format as the applied data exchange standard.

According to the control method of computing nodes applied to water conservancy construction sites according to the embodiments of the present invention, the central control node initiates regular polling, and the potential participating nodes report their respective remaining computing capabilities, and perform task decomposition according to the data reported by each node, Assign to designated nodes for calculation, and report the calculation results, and finally summarize the final results based on the information reported by each node. Therefore, this method makes full use of the remaining computing power of each computing node (such as sensors and data processing units) on the site, and can effectively improve the informatization level of the water conservancy construction site.

A further embodiment of the present invention also provides a control system applied to computing nodes in water conservancy construction sites. As shown in FIG. 5 , a control system 500 applied to computing nodes at a water conservancy construction site according to an embodiment of the present invention includes: a discovery module 510 , an allocation module 520 , a reporting module 530 and a control module 540 .

Wherein, the discovery module 510 is configured to discover multiple computing nodes that can be used for computing tasks through periodic polling. In some examples, combined with what is shown in Figure 3, it is specifically summarized as the following steps:

Step 1: Send a polling request based on the compute node list and enable the waiting timer. In other words, the central control node (included in the control module 540) sends polling requests one by one according to the most recently used (Most Recently Used, MRU) node list in the computing node table maintained by the last calculation, and waits for the computing node (as denoted as Nn) response, and open the waiting timer at the same time, denoted as Tn.

In some examples, preferably, the node table is in the form of a linked list, and its format is as follows:

Linked list head, total number-→node ID, node addressing→next node...→linked list tail.

In addition, in some examples, the central control node maintains an information table, which records the situation of each computing node, including the correspondence between the logical number and the physical number. Correspondingly, for the computing node whose physical number fails, the Delete accordingly in the information table. In other words, the node information includes the correspondence between the logical number and the physical number. The logical number refers to the ID of the computing node, and the physical number refers to the actual representation of the computing node. For example, for the network card, it is the Mac number. Correspondingly, for the physical Compute nodes with invalid serial numbers are also deleted from the information table. In a specific example, the general message format is expressed in a table as shown in Table 1 below:

serial number marked illustrate 9. message number message type 10. source node number node number 11. destination node number destination node number 12. version type The version number of this message type 13. timestamp Stores the number of seconds elapsed since 12:00:00 on January 1, 2000 14. serial number An increasing serial number 15. message body The message content corresponding to the message type 16. check code Algorithm using validation

Table 1

The actual content of a message is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<version>1.0</version>

<src_node>1</src_node>

<dest_node>2</dest_node>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>no</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Further, the central control node maintains more information, such as the information model diagram shown in Figure 4, which includes the information content shown in Table 2 below:

Table 2

In some examples, preferably, the polling request sent is:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<src_node>1</src_node>

<dest_node>2</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>no</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

In this step, the waiting timer may be set to 10s, for example.

Step 2: Each computing node receives the polling request, estimates its current computing capability, and sends it to the control module 540 (the control module includes the central control node), specifically including:

M=N+P1+P2,

Among them, M is the current computing capability of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate in the past period of time, and P2 is the expected CPU occupancy rate in the future period of time.

Further, in some examples, the format of the sent response message is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_ACK</message_id>;

<src_node>2</src_node>

<dest_node>1</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>1</seq_no>

</head>

<body>

<power>234</power>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Step 3: Before the waiting timer expires, the control module 540 judges whether multiple computing nodes can complete computing tasks according to the current computing capabilities of each computing node.

Step 4: If possible, use multiple nodes to complete the computing task, otherwise continue to send polling requests. That is, the central control node judges whether these computing nodes participating in this computing task (in consideration of redundancy) can complete the computing task according to the situation reported by each computing node. If it can be completed, it enters the assignment stage and allocates computing tasks. If it cannot be completed, the polling is continued to enter a larger scope discovery stage, and the point-to-point message in step 1 is changed to a broadcast polling request message. In some examples, the specific message format is as follows:

<? xmlversion="1.0"encoding="utf-8"? >

<message>

<head>

<message_id>DISCOVERY_REQ</message_id>;

<src_node>1</src_node>

<dest_node>2</dest_node>

<version>1.0</version>

<time_tamp>2013-12-13:00:03:45:234</time_tamp>

<seq_no>11</seq_no>

</head>

<body>

<broadcast>yes</broadcast>

</body>

<tail>

<checksum>123</checksum>

</tail>

</message>

Step 5: When the waiting timer expires, the discovery module 510 no longer waits for the response from the computing node, and discards the timed-out response message.

The allocating module 520 is used to respectively acquire the current computing capabilities of multiple computing nodes, decompose the computing tasks, and process the decomposed computing tasks collaboratively through multiple computing nodes. In some examples, this is outlined in the following steps:

Step A: Set the number of computing nodes as N, and decompose the computing task into m subtasks, where N>M. For example, multiple computing nodes are respectively N1, N2, ... Nn, and the computing task T is decomposed into m subtasks respectively T1, T2, ... Tm, and N>M, then the computing nodes N1, N2, ... Nn respectively process the subtasks correspondingly. Tasks T1, T2, ... Tm.

In some examples, the expression of the subtasks in this step is semantic, so that in a heterogeneous environment, it has nothing to do with the operating system where the computing nodes are located, and the above decomposed subtasks are uniformly described by MathML.

Step B: Send each subtask to the corresponding computing node, and start the timeout timer.

Step C: Maintain the packet liveness (heartbeat) timer H of each computing node, and regularly determine whether each computing node is invalid.

Step D: Receive the calculation results of each calculation node before the timeout timer expires.

It should be noted that, in the above process, for the purpose of fault tolerance, the target computing node simply discards the packets that do not match its own ID.

The reporting module 530 is used to report the processing result of each computing node. Specifically, the reporting module 530 reports the processing result of each computing node to the central control node, that is, to the control module 540 .

The control module 540 analyzes the processing results of each computing node to control multiple computing nodes.

To sum up, the system 500 of the present invention makes full use of the remaining computing power of a large number of various sensors and data processing units (that is, computing nodes) laid on the site, and forms a network through a suitable protocol, and the protocol can be mainly summarized as four The stages are: discovery, assignment, reporting and summary, as shown in Figure 2.

Specifically, in the discovery phase, the central control node initiates periodic polling to discover computing nodes that can be used for this computing task; in the assignment phase (assignment phase), it assigns the computing task Decomposition and collaborative calculation at the same time; the reporting stage is that the calculation nodes assigned sub-tasks report the calculation results to the central control node through the network. Among them, there are two kinds of reported results, one is calculation success, and the other is failure. Of course, due to It is carried out through the network, and the central control node is required to detect the failed computing node. If the assigned subtask cannot be completed within the specified time, the computing node will not report; the summary stage is to analyze and summarize the above reported results. The result is sent to another effective computing node, while waiting for the calculation result, the specific steps are similar to the distribution stage, which belongs to the secondary distribution. In even more extreme cases, the second allocation will not yield good results, and you can try multiple times until the set time is reached or the number of attempts is reached, and the calculation task is invalidated.

In some examples, in consideration of node failure, the allocation module 520 adopts a redundancy strategy, and the same decomposed subtask can be allocated to multiple computer nodes. This way, it is also possible to compare the calculation results of compute nodes performing the same task.

In one embodiment of the present invention, the communication protocol between computing nodes adopts XML (Extensible Markup Language, Extensible Markup Language) format and is standardized within large enterprises or follows corresponding international national standards. Specifically, as far as the existing technology is concerned, in the large-scale multi-unit distributed system of an enterprise, the usual method for data exchange between application units is generally through the network socket (Socket) protocol, and basically a custom report is adopted. However, due to the lack of uniform standards, these self-defined information formats are arbitrary, versatile, and inflexible, and cannot meet the reality of long IT construction cycles and emerging new technologies in enterprises. demand. Therefore, the present invention adopts a standard communication protocol in XML format as the applied data exchange standard.

According to the control system of the computing nodes applied to the water conservancy construction site according to the embodiment of the present invention, the central control node initiates regular polling, and the potential participating nodes report their respective remaining computing capabilities, and perform task decomposition according to the data reported by each node, Assign to designated nodes for calculation, and report the calculation results, and finally summarize the final results based on the information reported by each node. Therefore, the system makes full use of the remaining computing power of each computing node (such as sensors and data processing units) on the site, which can effectively improve the informatization level of the water conservancy construction site.

In describing the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", " Back", "Left", "Right", "Vertical", "Horizontal", "Top", "Bottom", "Inner", "Outer", "Clockwise", "Counterclockwise", "Axial", The orientation or positional relationship indicated by "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying the referred device or element Must be in a particular orientation, be constructed in a particular orientation, and operate in a particular orientation, and therefore should not be construed as limiting the invention.

In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of these features. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

In the present invention, unless otherwise clearly specified and limited, terms such as "installation", "connection", "connection" and "fixation" should be understood in a broad sense, for example, it can be a fixed connection or a detachable connection , or integrated; it may be mechanically connected or electrically connected; it may be directly connected or indirectly connected through an intermediary, and it may be the internal communication of two components or the interaction relationship between two components, unless otherwise specified limit. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention according to specific situations.

In the present invention, unless otherwise clearly specified and limited, the first feature may be in direct contact with the first feature or the first and second feature may be in direct contact with the second feature through an intermediary. touch. Moreover, "above", "above" and "above" the first feature on the second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. "Below", "beneath" and "beneath" the first feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is less horizontally than the second feature.

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" mean that specific features described in connection with the embodiment or example , structure, material or characteristic is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples and features of different embodiments or examples described in this specification without conflicting with each other.

Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention, those skilled in the art can make the above-mentioned The embodiments are subject to changes, modifications, substitutions and variations.

Claims (10)

1. A control method applied to a calculation node of a water conservancy construction site, characterized in that, comprising the following steps: Use periodic polling to discover multiple compute nodes available for computing tasks; Obtaining the current computing capabilities of the multiple computing nodes respectively, decomposing the computing tasks, and cooperatively processing the decomposed computing tasks through the multiple computing nodes; Each computing node sends the processing result to the central control node; The central control node analyzes the processing results of each computing node to control the multiple computing nodes. 2. The method for controlling computing nodes applied to a water conservancy construction site according to claim 1, wherein said periodic polling is used to find a plurality of computing nodes that can be used for computing tasks, specifically comprising: Send a polling request and enable the waiting timer according to the list of computing nodes; Each computing node receives the polling request, estimates its current computing capability, and sends it to the central control node, specifically including: M=N+P1+P2, Among them, M is the current computing power of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate in the past period of time, and P2 is the expected CPU occupancy rate in the future period of time; Before the waiting timer expires, the central control node judges whether multiple computing nodes can complete computing tasks according to the current computing capabilities of each computing node; If possible, use the plurality of nodes to complete the computing task, otherwise continue to send a polling request; When the waiting timer expires, no longer wait for the response from the computing node, and discard the timed-out response message. 3. The method for controlling computing nodes applied to water conservancy construction sites according to claim 1, wherein the current computing capabilities of the plurality of computing nodes are acquired respectively, and the computing tasks are decomposed, and the The multiple computing nodes cooperatively process the decomposed computing tasks, specifically including: Assuming that the number of computing nodes is N, and decomposing the computing task into m subtasks, where N>M; Send each subtask to the corresponding computing node and start the timeout timer; Regularly judge whether each computing node is invalid; Before the timeout timer expires, the calculation results of each calculation node are received. 4. the control method applied to the calculation node of the water conservancy construction site according to claim 3, is characterized in that, also comprises: With a redundant strategy, the subtasks of the same decomposition can be assigned to multiple computing nodes. 5. The method for controlling computing nodes applied to water conservancy construction sites according to any one of claims 1-4, wherein a communication protocol in XML format is adopted between the computing nodes. 6. A control system applied to computing nodes at water conservancy construction sites, characterized in that it comprises: A discovery module, the discovery module is used to discover multiple computing nodes that can be used for computing tasks through periodic polling; An allocation module, the allocation module is configured to respectively acquire the current computing capabilities of the multiple computing nodes, decompose the computing tasks, and cooperatively process the decomposed computing tasks through the multiple computing nodes; A reporting module, the reporting module is used to report the processing results of each computing node; A control module, the control module analyzes the processing result of each computing node to control the multiple computing nodes. 7. The control system applied to computing nodes at water conservancy construction sites according to claim 6, wherein the discovery module discovers a plurality of computing nodes that can be used for computing tasks through regular polling, specifically comprising: Send a polling request and enable the waiting timer according to the list of computing nodes; Each computing node receives the polling request, estimates its current computing capability, and sends it to the control module, specifically including: M=N+P1+P2, Among them, M is the current computing power of the computing node, N is the current CPU occupancy rate, P1 is the CPU occupancy rate in the past period of time, and P2 is the expected CPU occupancy rate in the future period of time; Before the waiting timer expires, the control module judges whether multiple computing nodes can complete computing tasks according to the current computing capabilities of each computing node; If possible, use the plurality of nodes to complete the computing task, otherwise continue to send a polling request; When the waiting timer expires, the discovery module no longer waits for a response from the computing node, and discards the timed-out response message. 8. The control system applied to computing nodes at water conservancy construction sites according to claim 6, wherein the distribution module obtains the current computing capabilities of the plurality of computing nodes respectively, and decomposes the computing tasks, And the decomposed computing tasks are collaboratively processed through the multiple computing nodes, specifically including: Assuming that the number of computing nodes is N, and decomposing the computing task into m subtasks, where N>M; Send each subtask to the corresponding computing node and start the timeout timer; Regularly judge whether each computing node is invalid; Before the timeout timer expires, the calculation results of each calculation node are received. 9. The control system applied to computing nodes at water conservancy construction sites according to claim 8, wherein the distribution module is also used to adopt a redundant strategy, and the same decomposed subtasks can be assigned to multiple computing nodes . 10. The control system for computing nodes applied to water conservancy construction sites according to any one of claims 6-9, wherein a communication protocol in XML format is used between the computing nodes.