CN104243579A - Computational node control method and system applied to water conservancy construction site - Google Patents
Computational node control method and system applied to water conservancy construction site Download PDFInfo
- Publication number
- CN104243579A CN104243579A CN201410465692.7A CN201410465692A CN104243579A CN 104243579 A CN104243579 A CN 104243579A CN 201410465692 A CN201410465692 A CN 201410465692A CN 104243579 A CN104243579 A CN 104243579A
- Authority
- CN
- China
- Prior art keywords
- computing node
- node
- computing
- calculation task
- water conservancy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Telephonic Communication Services (AREA)
Abstract
The invention provides a computational node control method applied to a water conservancy construction site. The computational node control method applied to the water conservancy construction site comprises the following steps that regular polling is used for discovering multiple computational nodes capable of being used for a computing task, the current computing capacity of each computational node is obtained, the computing task is decomposed, the decomposed computing task is processed through cooperation of the computational nodes, each computational node sends a processing result to a central control node, and the central control node analyzes the processing result of each computational node so as to control the computational nodes. According to the computational node control method applied to the water conservancy construction site, the residual computing capacity of the computational nodes (such as sensors and data processing units) of the construction site are utilized fully, and the informationalized level of the water conservancy construction site can be improved effectively. The invention further provides a computational node control system applied to the water conservancy construction site.
Description
Technical field
The present invention relates to distributed computing technology field, particularly a kind of control method and system being applied to the computing node at water conservancy construction scene.
Background technology
Along with Internet of Things and the quick of sensor network are popularized, the situation of sensor network at the construction field (site), is used to get more and more.These sensor networks are widely used in the various and business related aspects such as collecting temperature, humidity, pressure, personnel location information, and along with management is to digitlization, Informatization Development, are also that introducing and the development of other business lays a solid foundation.But for a long time, Each performs its own functions for each network and node thereof, mutually isolate, the object of general fit calculation and information fusion cannot be reached, than if any CPU adopts is the modern CPU of 32, but for a long time occupancy below 1%, calculating potentiality cannot be played far away, and central server works overloadingly when running into the operation of large calculation task and computation-intensive, computing cycle is long, affects Real time Efficiency decision-making.
But the solution at present about the problems referred to above is few, and what have just refer to some plain designs, and have also just propose Some Opinions, the inreal complete scheme that can be used to develop a distribution and calculate.
Summary of the invention
The present invention is intended to solve one of technical problem in above-mentioned correlation technique at least to a certain extent.
For this reason, one object of the present invention is to propose a kind of control method being applied to the computing node at water conservancy construction scene, the method takes full advantage of the residue computing capability of each computing node in job site (as transducer and data processing unit), effectively can promote water conservancy construction field data level.
Another object of the present invention is to provide a kind of control system being applied to the computing node at water conservancy construction scene.
To achieve these goals, the embodiment of first aspect present invention proposes a kind of control method being applied to the computing node at water conservancy construction scene, comprises the following steps: adopt periodic polling to find can be used for multiple computing nodes of calculation task; Obtain the computing capability that described multiple computing node is current respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment; Result is sent to center control nodes by each computing node respectively; Described center control nodes analyzes the result of each computing node to control described multiple computing node.
In addition, the control method being applied to the computing node at water conservancy construction scene according to the above embodiment of the present invention can also have following additional technical characteristic:
In some instances, described employing periodic polling, to find the multiple computing nodes that can be used for calculation task, specifically comprises: send polling request according to computing node list and enable waiting timer; Each computing node receives described polling request, estimates computing capability current separately, and is sent to center control nodes, specifically comprise:
M＝N+P1+P2，
Wherein, M is the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future; Before described waiting timer expires, the computing capability that described center control nodes is current according to each computing node, judges whether multiple computing node can complete calculation task; If of course, then described multiple node is used for described calculation task, otherwise has continued to send polling request; When described waiting timer expires, no longer wait for the response of computing node, and abandon receiveing the response of time-out.
In some instances, describedly obtain the current computing capability of described multiple computing node respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment, specifically comprise: set described multiple computing node as N number of, and described calculation task is decomposed into m subtask, wherein N>M; Each subtask is sent to corresponding computing node, and start overtime timer; Timing judges whether each computing node lost efficacy; Before described overtime timer expires, receive the result of calculation of each computing node.
In some instances, also comprise: the strategy taking redundancy, the subtask of same decomposition can be assigned to multiple computing node.
In some instances, the communication protocol of XML format is adopted between each computing node described.
According to the control method being applied to the computing node at water conservancy construction scene of the embodiment of the present invention, periodic polling is initiated by center control nodes, respective residue computing capability is reported by potential participation node, and according to the data that each node reports, carry out Task-decomposing, be assigned to specified node to calculate, and report result of calculation, the information finally reported according to each node is aggregated into final result.Therefore, the method takes full advantage of the residue computing capability of on-the-spot each computing node (as transducer and data processing unit), effectively can promote water conservancy construction field data level.
The embodiment of second aspect present invention provides a kind of control system being applied to the computing node at water conservancy construction scene, comprising: find module, and described discovery module is used for by periodic polling to find can be used for multiple computing nodes of calculation task; Distribution module, described distribution module is used for obtaining the current computing capability of described multiple computing node respectively, and is decomposed by described calculation task, and the calculation task after being decomposed by described multiple computing node associated treatment; Reporting module, described reporting module is for transmitting the result of each computing node; Control module, described control module analyzes the result of each computing node to control described multiple computing node.
In addition, the control system being applied to the computing node at water conservancy construction scene according to the above embodiment of the present invention can also have following additional technical characteristic:
In some instances, described discovery module, is specifically comprised: send polling request according to computing node list and enable waiting timer to find the multiple computing nodes that can be used for calculation task by periodic polling; Each computing node receives described polling request, estimates computing capability current separately, and is sent to control module, specifically comprise:
M＝N+P1+P2，
Wherein, M is the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future; Before described waiting timer expires, the computing capability that described control module is current according to each computing node, judges whether multiple computing node can complete calculation task; If of course, then described multiple node is used for described calculation task, otherwise has continued to send polling request; When described waiting timer expires, then described discovery module no longer waits for the response of computing node, and abandons receiveing the response of time-out.
In some instances, described distribution module obtains the current computing capability of described multiple computing node respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment, specifically comprise: set described multiple computing node as N number of, and described calculation task is decomposed into m subtask, wherein N>M; Each subtask is sent to corresponding computing node, and start overtime timer; Timing judges whether each computing node lost efficacy; Before described overtime timer expires, receive the result of calculation of each computing node.
In some instances, described distribution module is also for taking the strategy of redundancy, and the subtask of same decomposition can be assigned to multiple computing node.
In some instances, the communication protocol of XML format is adopted between each computing node described.
According to the control system being applied to the computing node at water conservancy construction scene of the embodiment of the present invention, periodic polling is initiated by center control nodes, respective residue computing capability is reported by potential participation node, and according to the data that each node reports, carry out Task-decomposing, be assigned to specified node to calculate, and report result of calculation, the information finally reported according to each node is aggregated into final result.Therefore, this system takes full advantage of the residue computing capability of on-the-spot individual computing node (as transducer and data processing unit), effectively can promote water conservancy construction field data level.
Additional aspect of the present invention and advantage will part provide in the following description, and part will become obvious from the following description, or be recognized by practice of the present invention.
Accompanying drawing explanation
Above-mentioned and/or additional aspect of the present invention and advantage will become obvious and easy understand from accompanying drawing below combining to the description of embodiment, wherein:
Fig. 1 is the flow chart of the control method of the computing node being applied to water conservancy construction scene according to an embodiment of the invention;
Fig. 2 is the four-stage schematic diagram that the control method of the computing node being applied to water conservancy construction scene according to an embodiment of the invention realizes;
Fig. 3 is the schematic diagram of discovery phase according to an embodiment of the invention;
Fig. 4 is the maintenance information model schematic of center control nodes according to an embodiment of the invention; And
Fig. 5 is the structured flowchart of the control system of the computing node being applied to water conservancy construction scene according to an embodiment of the invention.
Embodiment
Be described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Being exemplary below by the embodiment be described with reference to the drawings, only for explaining the present invention, and can not limitation of the present invention being interpreted as.
The control method and the system that are applied to the computing node at water conservancy construction scene according to the embodiment of the present invention are described below in conjunction with accompanying drawing.
Fig. 1 is the flow chart of the control method of the computing node being applied to water conservancy construction scene according to an embodiment of the invention.As shown in Figure 1, be applied to the control method of the computing node at water conservancy construction scene according to an embodiment of the invention, comprise the following steps:
Step S101, adopts periodic polling to find can be used for multiple computing nodes of calculation task.
Specifically, in some instances, shown in composition graphs 3, this step specifically comprises:
Step 1: send polling request according to computing node list and enable waiting timer.In other words, i.e. (the Most Recently Used of the computing node table safeguarded according to last computation of center control nodes the most multiplex mistake recently, MRU) node listing, successively send polling request, and wait for that computing node (as being designated as Nn) is responded, and open waiting timer simultaneously, be designated as Tn.
In some instances, preferably, node table have employed the form of chained list, and its form is as follows:
Linked list head, sum-→ node ID, node address → next node ... → chained list tail.
In addition, in some instances, center control nodes safeguards an information table, this table record situation of each computing node, contains the corresponding relation of logical number and physics numbering, correspondingly, for the computing node that physics numbering lost efficacy, corresponding deletion also will be done in information table.In other words, the corresponding relation of logical number and physics numbering is contained in nodal information, logical number refers to the ID of computing node, physics numbering refers to the expression of computing node reality, such as network interface card, it is No. Mac, accordingly, for the computing node that physics numbering lost efficacy, also in information table, do corresponding deletion.In concrete example, general message format form is expressed as follows shown in table 1:
| Sequence number | Indicate | Explanation |
| 1. | Message number | Type of message |
| 2. | Source node number | Node number |
| 3. | Destination node number | Destination node number |
| 4. | Release type | The version number of this type of message |
| 5. | Timestamp | Deposit the number of seconds passed after the 12:00:00 moment on January 1st, 2000 |
| 6. | Sequence number | The serial number number increased |
| 7. | Message body | The message content corresponding with type of message |
| 8. | Check code | Use the algorithm of verification |
Table 1
And the actual message content realized 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, center control nodes safeguards more information, and information model figure as shown in Figure 4, comprises the information content as shown in Table 2 below:
Table 2
In some instances, preferably, the polling request of transmission 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>
And in this step, waiting timer such as can be set to 10s.
Step 2: each computing node receives polling request, estimates computing capability current separately, and is sent to center control nodes, specifically comprise:
M＝N+P1+P2，
Wherein, M represents the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future.
Further, in some instances, the back message form of transmission 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 waiting timer expires, the computing capability that center control nodes is current according to each computing node, judges whether multiple computing node can complete calculation task.
Step 4: if passable, be then used for calculation task by multiple node, otherwise continue to send polling request.I.e. center control nodes situation about reporting according to each computing node, judge that can these computing nodes (when considering redundancy) participating in this calculation task complete calculation task, if can complete, then enter the appointment stage, namely perform step S103.If can not complete, then proceed poll, to enter wider discovery phase, and the point-to-point message in step 1 is changed to the PollingRequest message of broadcast.In some instances, message format is as follows particularly:
<？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 waiting timer expires, no longer waits for the response of computing node, and abandons receiveing the response of time-out.
Step S102, obtains the computing capability that multiple computing node is current respectively, and is decomposed by calculation task, and the calculation task after being decomposed by multiple computing node associated treatment.
Specifically, this step specifically comprises:
Steps A: set multiple computing node as N number of, and calculation task is decomposed into m subtask, wherein N>M.Such as multiple computing node is respectively N1, N2 ... Nn, calculation task T are decomposed into m subtask and are respectively T1, T2 ... Tm, and N>M, then computing node N1, N2 ... Nn processes subtask T1, T2 respectively accordingly ... Tm.
In some instances, the expression of the subtask in this step is semantization, and like this in isomerous environment, with the operating system independent at computing node place, and the unified MathML of employing in the subtask of above-mentioned decomposition describes.
Step B: each subtask is sent to corresponding computing node, and start overtime timer.
Step C: bag (heartbeat) timer H alive maintaining each computing node, and timing judges whether each computing node lost efficacy.
Step D: before overtime timer expires, receives the result of calculation of each computing node.
It should be noted that, in above process, for fault-tolerant object, target computing nodes, for receiving the unmatched message with self ID, simply abandons.
Step S103, result is sent to center control nodes by each computing node respectively.
Step S104, center control nodes analyzes the result of each computing node to control multiple computing node.
In sum, method of the present invention makes full use of and is laid on on-the-spot a large amount of various transducer and the residue computing capability of data processing unit (i.e. computing node), networking is carried out by suitable agreement, and this agreement mainly can be summarized as four-stage, be respectively: find, assign, report and gather, as shown in Figure 2.
Specifically, at discovery phase, initiate periodic polling by center control nodes, to find the computing node that can be used for this calculation task; The appointment stage (allocated phase) is under the prerequisite grasping existing computing node ability, is decomposed by calculation task, simultaneously cooperated computing; The stage is reported to be exactly that result of calculation is reported to center control nodes by network by the computing node being respectively assigned with subtask, wherein, the result reported has two kinds, one is calculate successfully, another one is failure, certainly owing to being undertaken by network, needs center control nodes to detect the computing node lost efficacy, if can not complete the subtask of distribution in official hour, then this computing node does not report; Aggregation stages is that the above-mentioned result reported is carried out general analyzes, and the result for failure is sent to effective computing node in addition, and wait for result of calculation, concrete steps and allocated phase roughly the same, belong to secondary distribution simultaneously.When more extreme, the result that secondary distribution also can not get, can repeatedly attempt again, until arrive the time of setting or this calculation task that cancels after reaching number of attempt.
In some instances, when considering node failure, method of the present invention takes redundancy strategy, and the subtask of same decomposition can be assigned to multiple computer node.Like this, the result of calculation of the computing node performing same task can also be compared.
In one embodiment of the invention, communication protocol between each computing node adopts XML (ExtensibleMarkup Language, extend markup language) form and carries out standardization by large enterprise inside or follow corresponding International Country standard.Specifically, with regard to prior art, in the large-scale multiple unit compartment system of enterprise, general web socket (Socket) agreement of passing through is as the usual method of carrying out exchanges data between applying unit, what substantially take is self-defined message format, no matter be fixed length or separator, but, because these self-defining information formats lack unified standard, random large, versatility, flexibility is not enough, can not meet the demand of the reality that enterprise's IT construction cycle is grown and new technology emerges in an endless stream.Therefore, the present invention adopts the communication protocol of the XML format of standard to be used as the data exchange standard applied.
According to the control method being applied to the computing node at water conservancy construction scene of the embodiment of the present invention, periodic polling is initiated by center control nodes, respective residue computing capability is reported by potential participation node, and according to the data that each node reports, carry out Task-decomposing, be assigned to specified node to calculate, and report result of calculation, the information finally reported according to each node is aggregated into final result.Therefore, the method takes full advantage of the residue computing capability of on-the-spot each computing node (as transducer and data processing unit), effectively can promote water conservancy construction field data level.
Further embodiment of the present invention additionally provides a kind of control system being applied to the computing node at water conservancy construction scene.As shown in Figure 5, be applied to the control system 500 of the computing node at water conservancy construction scene according to an embodiment of the invention, comprise: find module 510, distribution module 520, reporting module 530 and control module 540.
Wherein, find that module 510 is for can be used for multiple computing nodes of calculation task with discovery by periodic polling.In some instances, shown in composition graphs 3, specifically following steps are summarised as:
Step 1: send polling request according to computing node list and enable waiting timer.In other words, i.e. (the Most Recently Used of the computing node table safeguarded according to last computation of center control nodes (being contained in control module 540) the most multiplex mistake recently, MRU) node listing, successively send polling request, and wait for that computing node (as being designated as Nn) is responded, and open waiting timer simultaneously, be designated as Tn.
In some instances, preferably, node table have employed the form of chained list, and its form is as follows:
Linked list head, sum-→ node ID, node address → next node ... → chained list tail.
In addition, in some instances, center control nodes safeguards an information table, this table record situation of each computing node, contains the corresponding relation of logical number and physics numbering, correspondingly, for the computing node that physics numbering lost efficacy, corresponding deletion also will be done in information table.In other words, the corresponding relation of logical number and physics numbering is contained in nodal information, logical number refers to the ID of computing node, physics numbering refers to the expression of computing node reality, such as network interface card, it is No. Mac, accordingly, for the computing node that physics numbering lost efficacy, also in information table, do corresponding deletion.In concrete example, general message format form is expressed as follows shown in table 1:
| Sequence number | Indicate | Explanation |
| 9. | Message number | Type of message |
| 10. | Source node number | Node number |
| 11. | Destination node number | Destination node number |
| 12. | Release type | The version number of this type of message |
| 13. | Timestamp | Deposit the number of seconds passed after the 12:00:00 moment on January 1st, 2000 |
| 14. | Sequence number | The serial number number increased |
| 15. | Message body | The message content corresponding with type of message |
| 16. | Check code | Use the algorithm of verification |
Table 1
And the actual message content realized 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, center control nodes safeguards more information, and information model figure as shown in Figure 4, comprises the information content as shown in Table 2 below:
Table 2
In some instances, preferably, the polling request of transmission 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>
And in this step, waiting timer such as can be set to 10s.
Step 2: each computing node receives polling request, estimates computing capability current separately, and is sent to control module 540 (control module comprises center control nodes), specifically comprise:
M＝N+P1+P2，
Wherein, M is the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future.
Further, in some instances, the back message form of transmission 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 waiting timer expires, the computing capability that control module 540 is current according to each computing node, judges whether multiple computing node can complete calculation task.
Step 4: if passable, be then used for calculation task by multiple node, otherwise continue to send polling request.I.e. center control nodes situation about reporting according to each computing node, judge that can these computing nodes (when considering redundancy) participating in this calculation task complete calculation task, if can complete, then enter the appointment stage, Distribution Calculation task.If can not complete, then proceed poll, to enter wider discovery phase, and the point-to-point message in step 1 is changed to the PollingRequest message of broadcast.In some instances, message format is as follows particularly:
<？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 waiting timer expires, then find that module 510 no longer waits for the response of computing node, and abandon receiveing the response of time-out.
Calculation task for obtaining the current computing capability of multiple computing node respectively, and decomposes by distribution module 520, and the calculation task after being decomposed by multiple computing node associated treatment.In some instances, specifically following steps are summarised as:
Steps A: set multiple computing node as N number of, and calculation task is decomposed into m subtask, wherein N>M.Such as multiple computing node is respectively N1, N2 ... Nn, calculation task T are decomposed into m subtask and are respectively T1, T2 ... Tm, and N>M, then computing node N1, N2 ... Nn processes subtask T1, T2 respectively accordingly ... Tm.
In some instances, the expression of the subtask in this step is semantization, and like this in isomerous environment, with the operating system independent at computing node place, and the unified MathML of employing in the subtask of above-mentioned decomposition describes.
Step B: each subtask is sent to corresponding computing node, and start overtime timer.
Step C: bag (heartbeat) timer H alive maintaining each computing node, and timing judges whether each computing node lost efficacy.
Step D: before overtime timer expires, receives the result of calculation of each computing node.
It should be noted that, in above process, for fault-tolerant object, target computing nodes, for receiving the unmatched message with self ID, simply abandons.
Reporting module 530 is for reporting the result of each computing node.Specifically, the result of each computing node is reported to center control nodes by reporting module 530, also namely reports to control module 540.
Control module 540 analyzes the result of each computing node to control multiple computing node.
In sum, system 500 of the present invention makes full use of and is laid on on-the-spot a large amount of various transducer and the residue computing capability of data processing unit (i.e. computing node), networking is carried out by suitable agreement, and this agreement mainly can be summarized as four-stage, be respectively: find, assign, report and gather, as shown in Figure 2.
Specifically, at discovery phase, center control nodes initiates periodic polling, to find the computing node that can be used for this calculation task; The appointment stage (allocated phase) is under the prerequisite grasping existing computing node ability, is decomposed by calculation task, simultaneously cooperated computing; The stage is reported to be exactly that result of calculation is reported to center control nodes by network by the computing node being respectively assigned with subtask, wherein, the result reported has two kinds, one is calculate successfully, another one is failure, certainly owing to being undertaken by network, needs center control nodes to detect the computing node lost efficacy, if can not complete the subtask of distribution in official hour, then this computing node does not report; Aggregation stages is that the above-mentioned result reported is carried out general analyzes, and the result for failure is sent to effective computing node in addition, and wait for result of calculation, concrete steps and allocated phase roughly the same, belong to secondary distribution simultaneously.When more extreme, the result that secondary distribution also can not get, can repeatedly attempt again, until arrive the time of setting or this calculation task that cancels after reaching number of attempt.
In some instances, when considering node failure, distribution module 520 takes redundancy strategy, and the subtask of same decomposition can be assigned to multiple computer node.Like this, the result of calculation of the computing node performing same task can also be compared.
In one embodiment of the invention, communication protocol between each computing node adopts XML (ExtensibleMarkup Language, extend markup language) form and carries out standardization by large enterprise inside or follow corresponding International Country standard.Specifically, with regard to prior art, in the large-scale multiple unit compartment system of enterprise, general web socket (Socket) agreement of passing through is as the usual method of carrying out exchanges data between applying unit, what substantially take is self-defined message format, no matter be fixed length or separator, but, because these self-defining information formats lack unified standard, random large, versatility, flexibility is not enough, can not meet the demand of the reality that enterprise's IT construction cycle is grown and new technology emerges in an endless stream.Therefore, the present invention adopts the communication protocol of the XML format of standard to be used as the data exchange standard applied.
According to the control system being applied to the computing node at water conservancy construction scene of the embodiment of the present invention, periodic polling is initiated by center control nodes, respective residue computing capability is reported by potential participation node, and according to the data that each node reports, carry out Task-decomposing, be assigned to specified node to calculate, and report result of calculation, the information finally reported according to each node is aggregated into final result.Therefore, this system takes full advantage of the residue computing capability of on-the-spot each computing node (as transducer and data processing unit), effectively can promote water conservancy construction field data level.
In describing the invention, it will be appreciated that, term " " center ", " longitudinal direction ", " transverse direction ", " length ", " width ", " thickness ", " on ", D score, " front ", " afterwards ", " left side ", " right side ", " vertically ", " level ", " top ", " end " " interior ", " outward ", " clockwise ", " counterclockwise ", " axis ", " radial direction ", orientation or the position relationship of the instruction such as " circumference " are based on orientation shown in the drawings or position relationship, only the present invention for convenience of description and simplified characterization, instead of indicate or imply that the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore limitation of the present invention can not be interpreted as.
In addition, term " first ", " second " only for describing object, and can not be interpreted as instruction or hint relative importance or imply the quantity indicating indicated technical characteristic.Thus, be limited with " first ", the feature of " second " can express or impliedly comprise at least one this feature.In describing the invention, the implication of " multiple " is at least two, such as two, three etc., unless otherwise expressly limited specifically.
In the present invention, unless otherwise clearly defined and limited, the term such as term " installation ", " being connected ", " connection ", " fixing " should be interpreted broadly, and such as, can be fixedly connected with, also can be removably connect, or integral; Can be mechanical connection, also can be electrical connection; Can be directly be connected, also indirectly can be connected by intermediary, can be the connection of two element internals or the interaction relationship of two elements, unless otherwise clear and definite restriction.For the ordinary skill in the art, above-mentioned term concrete meaning in the present invention can be understood as the case may be.
In the present invention, unless otherwise clearly defined and limited, fisrt feature second feature " on " or D score can be that the first and second features directly contact, or the first and second features are by intermediary indirect contact.And, fisrt feature second feature " on ", " top " and " above " but fisrt feature directly over second feature or oblique upper, or only represent that fisrt feature level height is higher than second feature.Fisrt feature second feature " under ", " below " and " below " can be fisrt feature immediately below second feature or tiltedly below, or only represent that fisrt feature level height is less than second feature.
In the description of this specification, specific features, structure, material or feature that the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to describe in conjunction with this embodiment or example are contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not must for be identical embodiment or example.And the specific features of description, structure, material or feature can combine in one or more embodiment in office or example in an appropriate manner.In addition, when not conflicting, the feature of the different embodiment described in this specification or example and different embodiment or example can carry out combining and combining by those skilled in the art.
Although illustrate and describe embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, and those of ordinary skill in the art can change above-described embodiment within the scope of the invention, revises, replace and modification.
Claims (10)
1. be applied to a control method for the computing node at water conservancy construction scene, it is characterized in that, comprise the following steps:
Adopt periodic polling to find can be used for multiple computing nodes of calculation task;
Obtain the computing capability that described multiple computing node is current respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment;
Result is sent to center control nodes by each computing node respectively;
Described center control nodes analyzes the result of each computing node to control described multiple computing node.
2. the control method being applied to the computing node at water conservancy construction scene according to claim 1, is characterized in that, described employing periodic polling, to find the multiple computing nodes that can be used for calculation task, specifically comprises:
Send polling request according to computing node list and enable waiting timer;
Each computing node receives described polling request, estimates computing capability current separately, and is sent to center control nodes, specifically comprise:
M＝N+P1+P2，
Wherein, M is the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future;
Before described waiting timer expires, the computing capability that described center control nodes is current according to each computing node, judges whether multiple computing node can complete calculation task;
If of course, then described multiple node is used for described calculation task, otherwise has continued to send polling request;
When described waiting timer expires, no longer wait for the response of computing node, and abandon receiveing the response of time-out.
3. the control method being applied to the computing node at water conservancy construction scene according to claim 1, it is characterized in that, describedly obtain the current computing capability of described multiple computing node respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment, specifically comprise:
If described multiple computing node is N number of, and described calculation task is decomposed into m subtask, wherein N>M;
Each subtask is sent to corresponding computing node, and start overtime timer;
Timing judges whether each computing node lost efficacy;
Before described overtime timer expires, receive the result of calculation of each computing node.
4. the control method being applied to the computing node at water conservancy construction scene according to claim 3, is characterized in that, also comprise:
Take the strategy of redundancy, the subtask of same decomposition can be assigned to multiple computing node.
5. the control method being applied to the computing node at water conservancy construction scene according to any one of claim 1-4, is characterized in that, adopts the communication protocol of XML format between each computing node described.
6. be applied to a control system for the computing node at water conservancy construction scene, it is characterized in that, comprising:
Find module, described discovery module is used for by periodic polling to find can be used for multiple computing nodes of calculation task;
Distribution module, described distribution module is used for obtaining the current computing capability of described multiple computing node respectively, and is decomposed by described calculation task, and the calculation task after being decomposed by described multiple computing node associated treatment;
Reporting module, described reporting module is for reporting the result of each computing node;
Control module, described control module analyzes the result of each computing node to control described multiple computing node.
7. the control system being applied to the computing node at water conservancy construction scene according to claim 6, is characterized in that, described discovery module, is specifically comprised to find the multiple computing nodes that can be used for calculation task by periodic polling:
Send polling request according to computing node list and enable waiting timer;
Each computing node receives described polling request, estimates computing capability current separately, and is sent to control module, specifically comprise:
M＝N+P1+P2，
Wherein, M is the current computing capability of computing node, and N is current C PU occupation rate, and P1 is the CPU occupation rate of the past period, and P2 is the CPU occupation rate of expection a period of time in future;
Before described waiting timer expires, the computing capability that described control module is current according to each computing node, judges whether multiple computing node can complete calculation task;
If of course, then described multiple node is used for described calculation task, otherwise has continued to send polling request;
When described waiting timer expires, then described discovery module no longer waits for the response of computing node, and abandons receiveing the response of time-out.
8. the control system being applied to the computing node at water conservancy construction scene according to claim 6, it is characterized in that, described distribution module obtains the current computing capability of described multiple computing node respectively, and described calculation task is decomposed, and the calculation task after being decomposed by described multiple computing node associated treatment, specifically comprise:
If described multiple computing node is N number of, and described calculation task is decomposed into m subtask, wherein N>M;
Each subtask is sent to corresponding computing node, and start overtime timer;
Timing judges whether each computing node lost efficacy;
Before described overtime timer expires, receive the result of calculation of each computing node.
9. the control system being applied to the computing node at water conservancy construction scene according to claim 8, is characterized in that, described distribution module is also for taking the strategy of redundancy, and the subtask of same decomposition can be assigned to multiple computing node.
10. the control system being applied to the computing node at water conservancy construction scene according to any one of claim 6-9, is characterized in that, adopts the communication protocol of XML format between each computing node described.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410465692.7A CN104243579A (en) | 2014-09-12 | 2014-09-12 | Computational node control method and system applied to water conservancy construction site |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410465692.7A CN104243579A (en) | 2014-09-12 | 2014-09-12 | Computational node control method and system applied to water conservancy construction site |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104243579A true CN104243579A (en) | 2014-12-24 |
Family
ID=52230907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410465692.7A Pending CN104243579A (en) | 2014-09-12 | 2014-09-12 | Computational node control method and system applied to water conservancy construction site |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104243579A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208136A (en) * | 2015-11-04 | 2015-12-30 | 苏州墨华高科信息技术有限公司 | Elasticity parallel CFD (computational fluid dynamics) cloud computing system |
| CN105787175A (en) * | 2016-02-25 | 2016-07-20 | 中国农业大学 | Water conservancy model cloud computing method and device based on model combination |
| CN108540568A (en) * | 2018-04-23 | 2018-09-14 | 移康智能科技（上海）股份有限公司 | Computing capability sharing method and smart machine |
| CN110839220A (en) * | 2019-10-28 | 2020-02-25 | 无锡职业技术学院 | Distributed computing method and system based on wireless ad hoc network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101072133A (en) * | 2007-05-23 | 2007-11-14 | 华中科技大学 | High-performance computing system based on peer-to-peer network |
| US20100241741A1 (en) * | 2005-01-31 | 2010-09-23 | Computer Associates Think, Inc. | Distributed computing system having hierarchical organization |
| CN102063327A (en) * | 2010-12-15 | 2011-05-18 | 中国科学院深圳先进技术研究院 | Application service scheduling method with power consumption consciousness for data center |
| CN102929718A (en) * | 2012-09-17 | 2013-02-13 | 江苏九章计算机科技有限公司 | Distributed GPU (graphics processing unit) computer system based on task scheduling |
-
2014
- 2014-09-12 CN CN201410465692.7A patent/CN104243579A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100241741A1 (en) * | 2005-01-31 | 2010-09-23 | Computer Associates Think, Inc. | Distributed computing system having hierarchical organization |
| CN101072133A (en) * | 2007-05-23 | 2007-11-14 | 华中科技大学 | High-performance computing system based on peer-to-peer network |
| CN102063327A (en) * | 2010-12-15 | 2011-05-18 | 中国科学院深圳先进技术研究院 | Application service scheduling method with power consumption consciousness for data center |
| CN102929718A (en) * | 2012-09-17 | 2013-02-13 | 江苏九章计算机科技有限公司 | Distributed GPU (graphics processing unit) computer system based on task scheduling |
Non-Patent Citations (1)
| Title |
|---|
| 胡敏: "《对几种典型分布式计算技术的比较》", 《电脑知识与技术》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208136A (en) * | 2015-11-04 | 2015-12-30 | 苏州墨华高科信息技术有限公司 | Elasticity parallel CFD (computational fluid dynamics) cloud computing system |
| CN105787175A (en) * | 2016-02-25 | 2016-07-20 | 中国农业大学 | Water conservancy model cloud computing method and device based on model combination |
| CN108540568A (en) * | 2018-04-23 | 2018-09-14 | 移康智能科技（上海）股份有限公司 | Computing capability sharing method and smart machine |
| CN108540568B (en) * | 2018-04-23 | 2021-06-01 | 移康智能科技（上海）股份有限公司 | Computing capacity sharing method and intelligent equipment |
| CN110839220A (en) * | 2019-10-28 | 2020-02-25 | 无锡职业技术学院 | Distributed computing method and system based on wireless ad hoc network |
| CN110839220B (en) * | 2019-10-28 | 2022-12-20 | 无锡职业技术学院 | Distributed computing method based on wireless ad hoc network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104598425B (en) | A kind of general multiprocessing parallel calculation method and system | |
| CN105656653B (en) | Increase method of network entry, the device and system of node in distributed coordination system newly | |
| CN101163105B (en) | Method of obtaining path information and path computing node | |
| CN104243579A (en) | Computational node control method and system applied to water conservancy construction site | |
| CN113033082B (en) | Decentralized computing force perception-based decentralised federal learning framework and modeling method | |
| CN103313031A (en) | Dynamic dispatching video recording system and method | |
| CN105808343B (en) | For the cluster resource control method in complicated production management system | |
| CN103475524A (en) | Layered networking method based on master-slave node communication network | |
| TWI770749B (en) | Inspection method and device | |
| WO2013185713A2 (en) | Industrial control system and management device | |
| CN103582065A (en) | Efficient ad hoc network routing method for wireless sensor network | |
| WO2014000698A1 (en) | Ip layer-based network topology identification method and device | |
| CN104363173A (en) | Forwarding router information generation method, controller and software-defined networking system | |
| CN107517227A (en) | Session implementation method and device for distributed consensus system | |
| CN104168281B (en) | Power line carrier ad hoc network, apparatus and system | |
| CN111064672A (en) | Cloud platform communication system, election method and resource scheduling management method | |
| CN114301809B (en) | Edge computing platform architecture | |
| CN103001880B (en) | Based on the flow lead method and apparatus of non-standard OSPF Routing Protocol | |
| CN103037483A (en) | Network server and method for connecting same and application server | |
| CN113765743A (en) | Intelligent gateway working state monitoring method | |
| CN102307152B (en) | Method and device for pushing inter-domain resources | |
| CN103677994B (en) | Distributed data processing system, device and method | |
| CN106330774A (en) | Method and device for microwave link to transmit business data | |
| CN104065754A (en) | Wireless sensor network based on P2P technology and construction method thereof | |
| WO2015000229A1 (en) | Heartbeat model-based method for detecting communication state of large rectifying apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141224 |