KR100422151B1 - Method for transmitting performance monitoring data between an administration unit and shelves - Google Patents

Abstract

The present invention avoids data transmission according to a transmission request of a management unit in a system having a management unit for controlling a plurality of shelves, and allows each self to determine a transmission time by itself so that the accumulated performance monitoring data can be transmitted at a set cycle. The present invention provides a method for transmitting performance monitoring data between management units, the method comprising: setting a reference time for each main function unit of each self at a predetermined time interval; The process of the main functional unit includes collecting a specific type of data from each unit within a set range until a reference time arrives; The process of the main function unit sets the time when the reference time arrives as the reference time, calculates the data transmission start time by adding the extraction time calculated in the set manner to the reference time, and collects the data collected at the calculated data transmission start time. Transmitting to the management unit; The management process includes receiving data transmitted from each shelf, and processing the data in a set manner to form performance monitoring data when data reception is successfully completed. In this way, the performance monitoring data is transmitted by increasing the efficiency of data transmission procedure and processor operation.

Description

Method for transmitting performance monitoring data between an administration unit and shelves

The present invention relates to the transmission of performance monitoring data (Performance Monitoring Data, or PM data) between a multi-self and a management unit, and more particularly, data transmission according to a transmission request of a management unit in a system having a management unit controlling a plurality of self. The present invention relates to a method of transmitting performance monitoring data between a multi-self and a management unit so that each self can determine its own transmission time and transmit accumulated performance monitoring data at set intervals.

In general, some of the equipment that makes up an exchange or transmission system includes a number of shelves.

In these systems, the functional modules are housed in properly designed units, and the motherboard is racked and self-structured to make them easy to mount or dismount. The rack on the motherboard consists of a number of shelves, allowing each unit to be mounted.

Each shelf in the system performs a specific function, and the association or signal exchange between the shelves is coordinated and controlled by the main processor mounted on the motherboard or a specific shelf.

It is expected that any self will malfunction during system operation. If the self-error is neglected, errors can accumulate in the process of continuously processing the data processed in the self, which may cause a situation such as a system error or a loss of a data frame.

Therefore, the management unit to check the operation status of each self every certain period. Data representing the operation state of the self is collected in the self, which is referred to as performance monitoring data.

Describe the performance monitoring data transfer procedure between multiple self and management.

1 is a block diagram of a system including a general multiple self and a management unit, FIG. 2 is a flowchart illustrating a transmission of a management unit when transmitting performance monitoring data according to the prior art, and FIG. Flowchart.

Referring to FIG. 1, the hardware configuration of a system having a plurality of shelves includes a plurality of racks 110 to 140, and a management unit 100 connected to each rack 110 to 140 to exchange signals. Each rack 110-140 includes a proper number of shelves.

Here, there are four racks in the system and three racks for each rack. Normally, the number of racks is in the [1,30] section and the number of shelves per rack is in the [1,4] section.

Each shelf will include a Main Function Unit (or MFU) with a central processing unit (CPU). The main function unit controls and manages various units planted in its own shelf. In-house software processing is carried out in accordance with a program that defines certain procedures, which are carried out by the main functional unit.

The management unit 100 is implemented as a workstation capable of executing an application program. To deliver the monitored results to the operator, a workstation with data input / output capabilities is applied. The workstation of the management unit 100 performs a monitoring procedure for each self according to the application program ported by the operator.

In the case of transferring performance monitoring data between a plurality of shelves and a management unit, the transfer procedure of the self is realized by a program implanted in the main functional unit, and the transmission procedure of the management unit 100 is realized by an application program of the workstation.

In the conventional method, one management unit 100 requests transmission of performance monitoring data for each self. The request for data transmission is issued at regular intervals according to the order scheduled in the management unit 100.

In other words, the workstation of the management unit 100 starts a processor according to an application program and counts the time with the processor, and when the counted time reaches a predetermined time, the main function unit of the self is connected to itself. Issuing a transfer request.

In each shelf, the main function unit accumulates information about the execution status of each unit in the self, and forms the performance monitoring data. When the data transmission request of the management unit 100 is received, the main function unit transmits the accumulated performance monitoring data to the processor of the management unit 100.

Transmission control protocol / Internet protocol (TCP / IP) is used as a communication protocol between self and management.

According to Fig. 2, the process driven by the management unit starts the transmission procedure after waiting until every start time. At this time, the process simply performs the function of counting time.

The start time indicates the threshold of the counted time. When this start time is reached, a series of procedures for issuing a transfer request are performed. If the transfer request from the processor is issued every 15 minutes for each self, the start time comes every 15 minutes. In this 15 minutes, all performance monitoring data should be transmitted (S210).

For example, the management process issues data transfer requests every 00 minutes, 15 minutes, 30 minutes, and 45 minutes starting from 00 minutes every hour. The operation at the time of arrival every 15 minutes is repeated.

When the counted time reaches the start time, the process of the manager obtains the number of activated self. Self in the inactive state does not have the monitoring data (S220).

Once the number of self is confirmed, the process calculates the schedule time. The schedule time is calculated according to the time required for issuing a data transmission request to each shelf and receiving data from the shelf and the number of activated shelves (S230).

If the number of active self is N, the schedule time is calculated by dividing N by 15 minutes, which is the time interval until the next data transmission request. However, in consideration of signal exchange and processing time of each processor, the time divided by N in the middle of 12 minutes may be applied to allow 15 minutes before and 1 minute after 15 minutes.

The estimated schedule time is allocated for every N selfs activated. The execution schedule of the management unit processor is configured by this allocated time system.

The processor then searches the configured schedule to determine whether there is a self to issue a transfer request. If the transfer request is issued for all of the activated self, the process returns to step S210 and waits until the next start time (S240).

If there is a shelf to issue a transfer request, the process delivers the transfer request of performance monitoring data to that shelf. After transmitting the transmission request to one self, the processor returns to step S240 to check whether there is a remaining self on the schedule, and performs a procedure corresponding thereto (S250).

3, the process performed in each self main function unit waits until a performance monitoring data transmission request is received from the management unit. During this waiting time, the main function unit collects and accumulates performance monitoring data on the performance status of each unit in the corresponding self (S310 to S320).

When a data transfer request is received, the process of the main function unit delivers the accumulated performance monitoring data to the management unit. The transmitted data is received by the process of the management unit, and the data received from each shelf is collected. The process of the main function unit that has transmitted data resumes accumulation of performance monitoring data while waiting for the next data transmission request to be received (S330).

As such, the manager must identify the number of activated shelves and consider a schedule to configure a schedule for each shelf.

However, the number of activated self in the system is variable. Management should reconfigure the schedule to collect performance monitoring data every 15 minutes. Therefore, the prior art has a problem that the process load is heavy due to the repetition of the same task such as scheduling.

The present invention has been made to solve the above-mentioned conventional problems, and an object of the present invention is to avoid data transmission according to a transmission request of a management unit in a system having a management unit for controlling a plurality of self and each self has a transmission time. The present invention provides a method of transmitting performance monitoring data between a multi-self and a management unit to determine the data rate by itself and transmit accumulated performance monitoring data at set intervals.

1 is a block diagram of a system including a typical multiple shelf and management unit.

2 is a transmission flowchart of a management unit when transmitting performance monitoring data according to the related art.

3 is a transmission flowchart of a self unit during performance monitoring data transmission according to the related art.

4 is a transmission flowchart of a management unit in a method of transmitting performance monitoring data between multiple self and a management unit according to an exemplary embodiment of the present invention.

5 is a flow chart of a self in a method of transmitting performance monitoring data between a multi self and a management unit according to an exemplary embodiment of the present invention.

In order to achieve the above object, the present invention provides a method of transmitting performance monitoring data between a multi-self and a management unit, the method comprising: setting a reference time that arrives at a predetermined time interval for each self-main unit; The process of the main functional unit includes collecting a specific type of data from each unit within a set range until the reference time arrives; The process of the main function unit sets the time point when the reference time arrives as the reference time, calculates the start time of data transmission by adding the extraction time calculated in the set manner to the reference time, and at the calculated start time of data transmission. Transmitting the collected data to a management unit; The management unit may include receiving data transmitted from each self, and processing the data in a set manner to form performance monitoring data when data reception is successfully completed.

The present invention according to this configuration allows the management unit to efficiently monitor various types of data for a plurality of shelves in the system. Data collected by management is representative of performance monitoring data, but this method can be applied to any type of data transmission.

In the present invention, the management unit does not have to have a schedule for issuing a data transmission request for each self. That is, the management unit maintains the reception standby state, and determines whether the reception state is good when the data transmitted by each self according to a certain scheme is received.

When data is successfully received from each shelf, the management unit may process the data received at each shelf in a predetermined manner so that the execution state of each shelf is monitored.

There are multiple racks in the system, and each rack has multiple shelves. Each rack and shelf knows its unique number in relation to the management and other racks or other shelves. Based on this unique number, each shelf performs data exchange with the management unit, and the sequence and data transmission order between the shelves may be determined.

In order for the process performed in each shelf to collect data to be transmitted successfully, the same reference time must be set between the process of management and the process of each shelf. The reference time should be applied uniformly to all self and management units in the system. The reference time is a time index with magnitude. For example, the reference time can be set as 15 minutes.

From the time point at which the time coefficient starts, it is determined whether the reference time has arrived by the process of the main functional unit. The reference time is the time when the reference time has elapsed since the time count was started.

The process of the main functional unit shall have a timer for time counting and shall have information on the reference time. After each reference time has elapsed and a reference time arrives, a new time coefficient is started to determine whether the next reference time has elapsed.

The process of the main functional unit collects the necessary data from the units within a certain range during the reference time.

The range in which the main function unit collects data may be the self to which it belongs, or it may be units in other shelves that are not belonging to but connected to exchange data. More broadly, data collection may be possible in units other than the self, provided that the conditions of the data exchange path are established. Therefore, the scope of collecting data to be transmitted to the management unit by the process of the main functional unit is not limited in the present invention.

Preferably, the main function unit collects performance monitoring data from each unit in the self. Then, the data collected by the process of the main function unit is collected from each unit in the self belonging to the main function unit to form the performance monitoring data indicating the execution status of the units in the self. This data is collected and accumulated until it is transmitted to the management unit, and is newly accumulated from the point of completion of each transmission.

The process of the main functional unit must estimate the start time of data transfer in order to transfer the collected data to the management. The start time of data transmission is determined by the reference time specified by the passage of the reference time and the extraction time calculated in the set manner.

The extraction time is a time index, which has two uses. One is the usage applied to the calculation of extraction time, the minimum value of the time interval existing between each self start time of data transmission, and the other is the usage applied to the calculation of data transmission time by the sum of extraction time and reference time. Calculation of (t [i + 1] -t [i]) when (i) is assigned to each shelf for any positive integer (i) and the extraction time for each shelf is defined as t [i] Instruct the result. This usage is explicit.

The calculation of the extraction time should take into account the number of shelves and racks. These values are set in advance by the operator or scheduled in advance by statistical techniques using appropriate systems. In general, the number of racks is fixed and unchanged in system operation, but the number of shelves per rack and the total number of racks in the entire system are variable. Preferably, the number of racks in the system is invariant, and the number of shelves per rack is set to be the maximum number or less of shelves that can be mounted in one rack.

If it is assumed that the number of racks is fixed, the extraction time may be calculated by considering only the division of the shelf, not considering the division of the rack, depending on the operation method. That is, after setting the number of shelves to transmit data to the management unit, the extraction time may be calculated by accumulating the value obtained by dividing the reference time by the number of shelves for each shelf.

Then, the start time of data transmission of each shelf is allocated by assigning a sequence to each shelf, calculating the extraction time by the progressive accumulation method according to the sequence, and then calculating the extraction time by adding the calculated extraction time to the reference time.

If the extraction time is calculated by considering the number of racks, the extraction time is divided into two sub-unit times. The divided extraction times are referred to as extraction time 1 and extraction time 2, respectively.

After calculating the number of racks to transmit data to the management unit and the number of shelves per rack, the extraction time 1 is calculated by accumulating the calculated value by dividing the reference time by the total number of racks for each rack. The extraction time 2 is calculated by accumulating the value obtained by dividing the extraction time 1 calculated for each rack by the predetermined number of shelves per rack for each shelf.

With respect to the extracted extraction time 1 and extraction time 2, the extraction time 1 is added to the reference time to calculate the data transmission start time of each rack, and the extraction time 2 calculated by the self of the corresponding rack at the start time of data transmission of each rack. Calculate the data transfer start time for each self.

In the case of calculating the data transfer time of each shelf in consideration of the number of racks, each shelf can arbitrarily select a time while the calculation time 2 does not exceed the extraction time 2 within the range of the data transfer time of the rack.

For example, if the number of shelves per rack is 25, the number of possible extraction times 2 is 25 if three shelves are mounted in one rack. If the possible extraction time 2 is defined as the vector t [25], each shelf in the rack is randomly selected from the total 25 extraction times 2, and its sequence number such as t [0], t [12], t [18]. Can be determined.

In this case, although the choice of extraction time 2 is limited within the range of self-determination per rack, some randomness is guaranteed. This randomness ensures that each self's main functional unit has the availability of process management.

When the data transmission start time is calculated, each self cycle transmits the data collected by the unit to the management unit at the calculated data transmission start time.

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings.

4 is a flowchart illustrating a transmission of the management data between the multi-self and the management unit according to an embodiment of the present invention, and FIG. 5 is a transmission method of the monitoring data between the multi-self and the management unit according to an embodiment of the present invention. It is self-transfer flow chart.

This embodiment is applied to a system including a general multiple self and a management unit, which will be described with reference to FIG.

According to FIG. 4, the workstation of the manager 100 monitors the transmission path by starting a process to perform a data transmission procedure. The started process is in a waiting state. Here, the data transmitted between the management unit 100 and the self is performance monitoring data (S410).

When the process receives the data through the transmission path, the process determines whether the data is successfully received. If data reception fails, the process may return to step S410 after performing a procedure such as requesting retransmission of the corresponding data (S420).

Protocols related to the process of transmitting and receiving data are well known. Preferably, TCP / IP can be applied as a protocol related to data transmission and reception.

When the data reception is successfully completed, the process of the manager 100 properly processes the data received from the self to form performance monitoring data. In order to receive the next data, the process returns to step S410 to maintain the data reception standby state (S430).

According to Fig. 5, the main function unit of each shelf waits for every reference time. The reference time can be set in real time by the operator or automatically loaded by the boot-up program upon each self initialization. The setting method of the reference time is not limited by the present invention (S510).

Here, the case where the reference time is set to 15 minutes is illustrated. In this case, one minute before and two minutes after 15 minutes are used in consideration of the processing time of the process, and the time of 12 minutes is used for data transmission.

The process of the functional unit can determine whether the reference time has elapsed by operating a timer. The standard time arrives after each reference time elapses. For example, if each shelf in the system started its first time count from the hour, then the baseline time is reached for each shelf in every rack every 00 minutes, 15 minutes, 30 minutes, and 45 minutes.

When the reference time arrives, the process of the main functional unit calculates the extraction time and adds the reference time to calculate the data transmission time (S520).

Information on the number of shelves and racks required for the calculation of the extraction time is given to the main functional units of each shelf. This information can be set at initialization or in some other way.

In general, the number of racks (110-140) in the system is often fixed invariant. Therefore, the number of racks can be seen to be fixed, here is an example of a rack M.

The number of shelves mounted in each rack is variable. Therefore, the number of self should be adequately estimated. Here, the maximum number of shelves that can be mounted in the rack is N, and the maximum number of shelves is set as the number of shelves per rack.

For example, if there are four racks 110 to 140 and the maximum number of shelves per rack is 30, M and N become 4 and 30, respectively. Depending on the operating environment of the system, N may be set to 30 or less if the number of shelves mounted in each rack is very small compared to the maximum number of shelves allowed.

In the following description, let the reference time be T0 and the start time of data transmission be T1.

Let TS0 be the extraction time that is calculated when all the shelves are set without considering the number of racks. TS0 is a calculation result of (12 minutes / (M * N)). If the sequence vector composed of each shelf is set so that the sequence of the shelf is incremented by 1 sequentially from shelf 1 of rack 1, S [i] (i is a positive integer), the start time of data transmission of the i-th shelf is ( T1 = T0 + 1 minute + TS0 * i).

Since the values of T0 and TS0 are fixed in one routine that is performed after the reference time arrives at the shelf, T1 in each shelf has an interval of TS0 according to the sequence rank (i) of the shelf.

When the data transmission start time T1 is calculated, the process of the main functional unit transmits the performance monitoring data to the management unit at the time point when the data transmission start time arrives through the time coefficient (S530).

If the data transfer fails, the process of the main function unit returns to step S520 to recalculate the data transfer start time by using a time not allocated to the other shelves, and retry to transfer its data at the recalculated time. do. If the data transmission is successful, the process returns to step S510 and collects performance monitoring data until the next reference time elapses (S540).

Next, another embodiment of the present invention will be described.

Here, the extraction time of each shelf is calculated in consideration of the number of racks. At this time, most of the steps, including the transmission procedure of the manager 100 is the same as the first embodiment. Therefore, this embodiment describes only those parts that differ from the first embodiment.

In the case of considering the number of racks, it is assumed that extraction time 1 is TS1 and extraction time 2 is TS2 when performing step S520.

TS1 is calculated by accumulating the value obtained by dividing the reference time T0 by the predetermined number of racks M for each rack, and is calculated as (TS1 = 12 minutes / M).

TS2 is calculated by accumulating the value obtained by dividing the extraction time 1 TS1 calculated for each rack by the number N of predetermined shelves per rack for each shelf, and is calculated as (TS2 = TS1 / N). TS2 has the same operation formula as TS0 when the number of racks is not taken into account.

All shelves mounted in each rack shall transmit all data during extraction time 1 (TS1) of the rack. Normally, N exceeds the number of actually mounted shelves, and each shelf can arbitrarily select a time to transmit data within the range of TS1.

The transmission start time (T2) of rack i is determined by the operation result of (T2 = T0 + 1 minute + TS1 * i), and any j of rack i (where j is not greater than TS1 / TS2) For each self, the transmission start time T3 is calculated as (T3 = T2 + TS2 * j).

For example, if the reference time is 13:00, rack 1 must complete the data transfer in [13: 1, 13:00 (1 + TS1) minutes]. Rack 2 starts data transmission at 13:00 (1 + TS1). This system determines the start time of data transfer for all shelves.

As described above, when data is transmitted between a plurality of self and the management unit, the self determines the data transmission time and transmits the data so that a conventional problem such as a processing load due to the repetitive scheduling task of the management unit does not occur.

Since the actual processing time is in seconds, the process load can be sufficiently distributed, and even if the data transmission performed within the set transmission time fails, the data transmission can be retried using the time of 2 minutes after it is left.

The embodiments described above are within the scope of various changes, modifications, and equivalents of the present invention. Therefore, the present invention is not limited to the description of the examples.

According to the method of transmitting performance monitoring data between the multiple shelves and the management unit of the present invention, there is an advantage of solving the load weighting problem of performing the data transmission schedule for a plurality of shelves in one management unit.

In addition, in the system to which the present invention is applied, each self determines its own data transmission start time to transmit performance monitoring data, thereby increasing the efficiency of the data transmission procedure and reducing the load on the management unit, thereby enabling efficient operation of the processor. It works.

Claims (7)

Setting a reference time for each self functioning unit at a predetermined time interval; The process of the main functional unit includes collecting a specific type of data from each unit within a set range until the reference time arrives; The process of the main function unit sets the time point when the reference time arrives as the reference time, calculates the start time of data transmission by adding the extraction time calculated in the set manner to the reference time, and at the calculated start time of data transmission. Transmitting the collected data to a management unit; The process of the management unit receives the data transmitted from each shelf, and if the data reception is successfully completed, the process comprising the step of processing the data in a set manner to form the performance monitoring data, characterized in that the performance monitoring between the multiple self and the management unit How to transfer data. The method of claim 1, wherein the data collected by the process of the main functional unit, Collected and collected from each unit in the self belonging to the main function unit to configure the performance monitoring data indicating the performance status of the units in the self, collect and accumulate until transmitted to the management unit, and is newly accumulated from the point of completion of each transmission Method of transmitting performance monitoring data between multiple self and management unit. The method of claim 1, wherein the calculation method of the extraction time, After the number of the self to transmit the data to the management unit is scheduled, the value calculated by dividing the reference time divided by the predetermined number of self for each self self-calculated, characterized in that to calculate the performance of the monitoring data between the multi-self and the management unit Transmission method. The method of claim 1, The calculation method of the extraction time is scheduled by the number of racks and the number of racks to be transmitted data to the management unit, and then the value calculated by dividing the reference time by the predetermined number of racks for each rack, extraction time 1 is The extraction time 1 is calculated by dividing the value calculated by dividing the extraction time 1 calculated for each rack by the predetermined number of shelves per rack, and calculating the extraction time 2, With respect to the calculated extraction time 1 and extraction time 2, the extraction time 1 is added to the reference time to calculate the start time of data transmission of each rack, and the self time of the corresponding rack is calculated at the start time of data transmission of each rack. And transmitting the extracted time 2 to calculate a start time of data transmission of each self. The method according to claim 3 or 4, Each shelf and each rack are arranged in sequence according to a predetermined sequence, and the extraction time is sequentially accumulated according to the arrangement order to be summed up to the reference time so that the data transfer start time is calculated. Method of transmitting monitoring data between management departments. The method according to any one of claims 1 to 4, wherein the data transmission start time is A margin is set for each of the lower limit and the upper limit of the reference time, and the lower limit is determined according to the result of adding up the margin time from the lower limit of the reference time to the reference time. And the upper limit is determined as a result value obtained by adding up the marginally advanced value from the upper limit to the reference time. The method according to any one of claims 1 to 4, wherein the data transmission start time is A first margin and a second margin are respectively provided at the lower limit and the upper limit of the reference time, so that the lower limit is determined according to the result value obtained by adding the value advanced up from the lower limit of the reference time by the first margin to the reference time. And the upper limit is determined as a result value obtained by adding up a value proceeding downward by the second margin from the upper limit of the reference time to the reference time.