Embodiment
For making the object of the invention, technical scheme and advantage clearer, below in conjunction with embodiment and accompanying drawing, to further explain of the present invention.
As process object, each network element of gathering from network element obtains the corresponding performance data of specified performance index sign, calculates the overall performance data by the performance data of obtaining again with system-level network element set in the present invention.Referring to Fig. 1, be the disposal route indicative flowchart of the system-level performance data of the present invention, this method may further comprise the steps:
Step 101, each network element that comprises from the network element set obtains the corresponding performance data of specified performance index sign.
Said network element set is the set that need obtain each network element of performance data.
Said obtaining can be specially: each network element that comprises to the network element set sends the collection request that comprises said specified performance index sign one by one, receives the performance data of each network element feedback.When certain network element sends the collection request, can start timer, if timer expiry does not also receive the performance data of this network element feedback, then abandon obtaining the performance data of this network element feedback.
Step 102 judges that the type of obtaining performance data is number percent type or numeric type, if the number percent type, then execution in step 103, if numeric type, then execution in step 104.
Step 103 is asked on average the performance data of obtaining, and the mean value that obtains as the overall performance data, and is added to the overall performance data in the performance table.
Step 104 is sued for peace the performance data of obtaining, and the summing value that obtains as the overall performance data, and is added to the overall performance data in the performance table.
Adopt the present invention program, can monitor in real time system-level performance data.Correspondingly, in Fig. 1 flow process, before the step 101, also comprise: when needs are monitored in real time, creation task, said task comprises each net element information, specified performance index and the monitoring frequency that the network element set comprises; Step 103 is said add to the overall performance data in the performance table after, also comprise: when waiting for that according to the monitoring frequency the next poll time arrives, return execution in step 101; Step 104 is said add to the overall performance data in the performance table after, also comprise: when waiting for that according to the monitoring frequency the next poll time arrives, return execution in step 101.
Adopt the present invention program, also can collect system-level History Performance Data.Correspondingly, in Fig. 1 flow process, before the step 101, also comprise:
In task list, add each task to be carried out, said task comprises each net element information, specified performance index, polling cycle and the correlation threshold that the network element set comprises, and also comprises the next poll time of corresponding each task in the said task list;
Judge the current task of need carry out poll that whether has, if then execution in step 101; Otherwise the minimum value of next poll time in the query task table is waited for when said minimum value arrives execution in step 101.The pairing task of minimum value of poll time next time just leaves nearest task of current time in the task list.
Adopt the present invention program, also can when carrying out the performance data processing, alarm.Correspondingly, in the flow process of Fig. 1, step 103 and step 104 are said add to the overall performance data in the performance table after, also comprise: whether judge the overall performance data greater than correlation threshold, if then send alarm to the webmaster sophisticated equipment.Said correlation threshold is used to weigh the overall performance data and whether surpasses normal value.Said webmaster sophisticated equipment is the equipment of receiving alarm in the network management system.
Through Fig. 2 and Fig. 3 the disposal route of the system-level performance data of the present invention is illustrated the process flow diagram of Fig. 2 for system-level performance data is monitored in real time, the process flow diagram of Fig. 3 below for system-level History Performance Data is collected.
The flow process of Fig. 2 may further comprise the steps:
Step 201 gets into system-level real-time monitoring interface, creation task.
Need the creation task content in system-level real-time monitoring interface input, comprise each net element information, specified performance index and monitoring frequency etc. that the network element set comprises; Wherein, Net element information comprises IP address and Simple Network Management Protocol (snmp, simple network management protocol) port numbers of network element etc., just can know the position of each network element through net element information; The specified performance index is need be to a certain performance index of this network element set monitoring; The specified performance index for example is memory usage or CPU usage etc., the frequency that the monitoring frequency representation is monitored this network element set, the just frequency of execution in step 202.
Step 202, each network element that comprises to the network element set sends the collection request that comprises specified performance index sign one by one.
Through the IP address and the snmp port numbers of network element, the collection request that will comprise specified performance index sign sends to corresponding network element.
Step 203 receives the performance data that each network element feeds back.
When certain network element sends the collection request, can start timer, if timer expiry does not also receive the performance data of this network element feedback, then abandon obtaining the performance data of this network element feedback.
Step 204 judges that the type of obtaining performance data is number percent type or numeric type, if the number percent type, then execution in step 205; If numeric type, then execution in step 206.
Step 205 is asked on average the performance data of obtaining, with the mean value that obtains as the overall performance data, execution in step 207.
Step 206 is sued for peace the performance data of obtaining, with the summing value that obtains as the overall performance data, execution in step 207.
Step 207 is added the overall performance data in the performance table to.
The real-time monitoring form that the said performance table of this step promptly obtains particularly, can be set to the time by horizontal ordinate, and ordinate is set to the overall performance data, and said indicators of overall performance for example is overall memory usage.
Step 208 when waiting for that according to the monitoring frequency the next poll time arrives, is returned execution in step 202.
Adopt Fig. 2 scheme, obtain the overall performance data of certain network element set at needs, the real-time creation task of webmaster personnel just can be known the overall performance data by the performance table that generates.
The flow process of Fig. 3 may further comprise the steps:
Step 301, interpolation task in task list, said task comprise each net element information, specified performance index, polling cycle and the correlation threshold that the network element set comprises, and also comprise the next poll time of corresponding each task in the said task list.
In task list, can add at least one task; Each task can be carried out performance data to the different specified performance indexs of identical network element set and collected; Also can carry out performance data and collect, also can carry out performance data and collect to the different specified performance indexs of different network elements set to the identical specified performance index of different network elements set.
The task definition of adding comprises each net element information, specified performance index, polling cycle and the correlation threshold that the network element set comprises; Net element information comprises IP address and snmp port numbers of network element etc.; Just can know the position of each network element through net element information; The specified performance index is for gathering a certain performance index of collecting to this network element; The specified performance index for example is memory usage or CPU usage etc., and polling cycle is represented the period that performance data is collected is carried out in this network element set, and correlation threshold is used to judge whether to alarm.The next poll time that also comprises corresponding each task in the said task list, when in task list, adding certain task, the next poll time of this task is for accomplishing the time of adding.
Step 302 is judged the current task of need carry out poll that whether has, if then execution in step 303, otherwise, execution in step 304.
Judge task of whether existing the next poll time to arrive in the task list,, this task is carried out poll if having.
Step 303, each network element that comprises to the network element set sends the collection request that comprises said specified performance index sign one by one.
Through the IP address and the snmp port numbers of network element, the collection request that will comprise specified performance index sign sends to corresponding network element.
Step 304, the minimum value of next poll time in the query task table is waited for when said minimum value arrives execution in step 303.
Step 305 receives the performance data that each network element feeds back.
When certain network element sends the collection request, can start timer, if timer expiry does not also receive the performance data of this network element feedback, then abandon obtaining the performance data of this network element feedback.
Step 306 judges that the type of obtaining performance data is number percent type or numeric type, if the number percent type, then execution in step 307; If numeric type, then execution in step 308.
Step 307 is asked on average the performance data of obtaining, with the mean value that obtains as the overall performance data, execution in step 309.
Step 308 is sued for peace the performance data of obtaining, with the summing value that obtains as the overall performance data, execution in step 309.
Step 309 is added the overall performance data in the performance table to.
The historical performance table that the said performance table of this step promptly obtains.
Whether step 310 judges the overall performance data greater than correlation threshold, if then send alarm to the webmaster sophisticated equipment.
If be not more than correlation threshold, then do not send alarm.
Through alarm, a certain performance index that the webmaster personnel can understand the network element set have occurred unusually, and make corresponding adjustment.
Step 311 is with the next poll time and the polling cycle summation of the current polling tasks that writes down in the task list, with the next poll time of the current polling tasks that writes down in the summing value updating task table that obtains.
Adopt Fig. 3 scheme, the webmaster personnel add in task list after each task, just can know the History Performance Data of each task automatically, and can know that the overall performance data have surpassed normal value through alarm.
Referring to Fig. 4, be the treatment facility structural representation of the system-level performance data of the present invention, this equipment comprises acquiring unit, judging unit and record cell;
Said acquiring unit is used for obtaining the corresponding performance data of specified performance index sign from each network element that the network element set comprises, and sends said judging unit to;
Said judging unit; Be used to judge that the type of obtaining performance data is number percent type or numeric type, if the number percent type then asks average to the performance data of obtaining; As the overall performance data, and the overall performance data are sent the mean value that obtains to said record cell; If numeric type then sues for peace the performance data of obtaining, as the overall performance data, and the overall performance data are sent the summing value that obtains to said record cell;
Said record cell is used to receive the overall performance data from said judging unit, adds in the performance table.
Alternatively, said acquiring unit comprises the collection subelement, is used for sending the collection request that comprises said specified performance index sign one by one to each network element that the network element set comprises, and receives the performance data of each network element feedback.
Alternatively, this equipment also comprises first poll judgement subelement and the first establishment subelement;
Said first creates subelement, is used for creation task, and said task comprises each net element information, specified performance index and the monitoring frequency that the network element set comprises;
Said record cell also is used for after adding the overall performance data to the performance table, sends the poll decision instruction to said first poll judgement subelement;
Said first poll judgement subelement; Be used to receive poll decision instruction from said record cell; Create subelement from said first and obtain the monitoring frequency, when waiting for that according to the monitoring frequency the next poll time arrives, send enabled instruction to said collection subelement;
Said collection subelement; Also be used to receive enabled instruction from said first poll judgement subelement; Create network element each net element information of set and the specified performance index that subelement obtains current task from said first; Each network element that comprises to the network element set sends the collection request that comprises said specified performance index sign one by one, receives the performance data of each network element feedback.
Alternatively, this equipment also comprises second poll judgement subelement and the second establishment subelement;
Said second creates subelement, is used in task list interpolation task, and said task comprises each net element information, specified performance index, polling cycle and the correlation threshold that the network element set comprises, and also comprises the next poll time of corresponding each task in the said task list;
Said record cell also is used for after adding the overall performance data to the performance table, sends the poll decision instruction to said second poll judgement subelement;
Said second poll judgement subelement is used to judge the current task of need carry out poll that whether has, if then send enabled instruction to said collection subelement; Otherwise the minimum value of next poll time in the query task table is waited for and is sent enabled instruction to said collection subelement when said minimum value arrives;
Said collection subelement; Also be used to receive enabled instruction from said second poll judgement subelement; Create network element each net element information of set and the specified performance index that subelement obtains current task from said second; Each network element that comprises to the network element set sends the collection request that comprises said specified performance index sign one by one, receives the performance data of each network element feedback.
Alternatively, this equipment also comprises Alarm Unit, and said record cell also is used for after adding the overall performance data to the performance table, sends the alarm that comprises the overall performance data to said Alarm Unit and differentiates indication;
Said Alarm Unit; Be used to receive from the alarm of said record cell and differentiate indication, create the correlation threshold that subelement obtains current task from said second, judge that whether the overall performance data are greater than correlation threshold; If then send alarm to the webmaster sophisticated equipment.
The present invention is applicable to the multiple network system, like the snmp protocol network system.
As process object, each network element of gathering from network element obtains the corresponding performance data of specified performance index sign, calculates the overall performance data by the performance data of obtaining again, adds in the performance table with system-level network element set in the present invention.Thereby, realize obtaining to system-level performance data.
The above is merely preferred embodiment of the present invention, and is in order to restriction the present invention, not all within spirit of the present invention and principle, any modification of being made, is equal to replacement, improvement etc., all should be included within the scope that the present invention protects.