JP2010128730A - Measurement data transfer control device, method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device etc. capable of efficiently performing data transfer when transferring measurement data to be observed to a subsequent stage so that a fatal loss of measurement information is prevented even if the measurement data exceeds a transfer band. <P>SOLUTION: The device includes: (A) a data receiver unit for successively receiving occurrence event information and for generating time information including an absolute elapsed time from the start of measurement; (B) a buffer for tentatively storing successive event information and time information; a data amount decision unit for deciding a data amount (C1) initially to be in a first stage, (C2) to be in a second stage when the buffer use rate exceeds a first ascending time threshold, and (C3) in the case of the second stage, to have returned to the first stage when the use rate falls below a first descending time threshold smaller than the first ascending time threshold; and a data transfer unit (D1) in the case of the first stage, for transferring the event information and the time information from the buffer to the subsequent stage intact, and (D2) in the case of the second stage, for transferring the event information from the buffer and coarse time information, having degraded accuracy of the time information, successively to the subsequent stage. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention generally relates to measurement data transfer control. In particular, the present invention relates to a transfer control system, a method therefor, and a computer program for transferring measurement data about real-time operation of a system or apparatus to be observed to a subsequent apparatus.

In recent years, the behavior of the system or device to be observed is measured in real time over a long period of time, and the measured data is recorded and analyzed to debug, test, improve performance, and reduce power consumption of the system or device. A method for performing the above is widely used.
FIG. 1 shows a system or device to be observed 102 and a “measuring device” 104 for measuring the behavior of these objects to be observed and transferring the measured data to the recording / analyzing device in the next stage, and receiving data from the measuring device. 1 shows an example of a measurement / recording / analysis environment of real-time behavior of an observation target, which comprises a “recording / analyzing device” 106 for recording / analyzing it.
The “measuring device” 104 is large and processes a large amount of data measured from the observation target, a data input interface 112 for connecting to the observation target 102, a data output interface 118 for connecting to the recording / analysis device 106. And a measurement data processing device 114. The measurement data processing device 114 includes an internal buffer 116 for temporarily storing processed data.

  The data sent to the recording / analyzing device 106 is information for grasping the behavior of the observation target. Specifically, it is composed of “event information” indicating the type of event that has occurred in the observation target, and “time information” indicating when the event has occurred.

Data measured via the data input interface 112 is temporarily stored (buffered) in the internal buffer 116 and transferred to the recording / analyzing device 106 at the subsequent stage via the data output interface 118. If the amount of data to be transferred does not exceed this transfer bandwidth, the internal buffer 116 will not overflow. However, when the amount of data measured is large, this transfer bandwidth is exceeded, so the frequency with which data to be transferred is taken out from the internal buffer must be reduced. As a result, data measured in the internal buffer is accumulated one after another, and the internal buffer eventually overflows. At this time, loss of overflowed data occurs.
In other words, the amount of data that can be measured in real-time measurement is limited to the transfer bandwidth when the measuring device transfers data to the recording / analyzing device in the subsequent stage. When information is measured, part of the information is lost.

In real-time measurement, the amount of data to be measured varies greatly depending on the operating status of the system even for the same observation target. For example, when a large number of measurement events occur in a short period of time.
For this reason, the amount of data is often found only after actual measurement. Even if the measurement device is designed before the measurement, it is difficult to determine the optimum bandwidth in advance, and it is only possible to design and implement the peak data amount.
However, if the peak data amount is estimated with a sufficient margin and the transfer bandwidth is designed, it becomes a costly design due to the implementation resources required for the peak time, which rarely occurs, and conversely If a sufficient margin for the amount of data is not taken, there is a high possibility that information will occur beyond the transfer bandwidth. In this case, a part of the measurement data will be lost, but it is only known that "a part of the measurement data has been lost", so the ideal transfer bandwidth that the measurement device should actually have Itself remains unknown. In other words, since the optimum transfer bandwidth is not known for recording and analyzing the information to be observed without excess or deficiency, the optimum reconstruction of the measurement, recording and analysis environment for the observation target cannot be performed accurately.

  Japanese Patent Application Laid-Open No. 08-130647 suggests a technique for reducing the amount of data by dynamically changing the data format according to the data characteristics. Japanese Patent Laid-Open No. 11-184733 uses a second data format in which certain information is reduced from the first data format in order to reduce the data amount of failure data. The technology for restoring data by adding data is disclosed.

  However, these techniques are based on a data transfer environment in which the transfer amount in transferring raw (normal) data is known. In other words, after a sufficient transfer bandwidth is secured to transfer the raw data, the data can be reduced and the data can be restored from the original data in order to reduce the transfer amount and perform efficient data transfer. This is a technique for transferring data by converting it into a data format in which a part of information or a part of information is deleted.

  Therefore, transfer the measured data to the recording / analyzing device for the observation target that is under development or testing so that it is impossible to predict how often the measured data will be obtained per unit time (ie, data volume). In this case, since the transfer data amount exceeds the transfer band, a part of the measurement data may be lost.

Japanese Patent Laid-Open No. 08-130647 Japanese Patent Laid-Open No. 11-184733

  The purpose of the present invention is when the measurement data about the real-time operation of the system or apparatus to be observed is transferred to the recording / analyzing apparatus in the subsequent stage, and the measurement data to be transferred increases as it exceeds the transfer band. Another object is to provide an apparatus and the like that can perform data transfer efficiently by adapting so that measurement information is not lost.

In order to solve the above-described problems, the present invention provides a measurement data processing apparatus that measures an event occurring in an observation target and transfers the measured data to the next stage.
In the first aspect of the present invention, the apparatus receives (A) event information indicating the type of the event that has occurred, and starts measurement related to the event information until the event occurs. A data receiving unit is provided for generating time information including the absolute elapsed time. The device also includes (B) an internal buffer that temporarily stores this event information and time information. The apparatus further includes (C) a data amount determining unit that determines the amount of data stored in the internal buffer, and the data amount determining unit is (1) in the first stage where the data amount is the lowest. In addition, when the usage rate of the data amount in the internal buffer exceeds the first rise threshold, it is determined that the data amount has become the second stage which is larger than the first stage, and (2) the second stage is reached. In some cases, if the usage rate falls below a first falling threshold value that is smaller than the first rising threshold value, it is determined that the amount of data has returned to the first stage. The apparatus further includes (D) a data transfer unit. (1) When it is determined that the data transfer unit is in the first stage, the event information and the time information are extracted from the internal buffer, and the extracted event When the information and time information are transferred to the next stage and (2) it is determined that the second stage, the event information and time information are taken out from the internal buffer, and the accuracy of the event information and time information is determined. The reduced coarse time information is transferred to the next stage.

  Further, according to a second aspect of the present invention, in the apparatus according to the first aspect of the present invention, when the data amount determination unit is further in the (4) second stage, the usage rate is at the time of the first increase. When the second rising threshold value, which is greater than the threshold value, is exceeded, it is determined that the data amount has reached the third stage, which is greater than the second stage. (5) When the third stage is in the third stage, the usage rate increases to the first level. When the value falls below the second falling threshold value that is larger than the hour threshold value and smaller than the second rising threshold value, it is determined that the data amount has returned to the second stage. Further, in the apparatus according to the first aspect of the present invention, when the data transfer unit further includes (3) a statistical recording unit and (4) it is determined that it is the third stage, the event information is read from the internal buffer. Time information is taken out, only this event information taken out is transferred to the next stage, and the statistical information generated based on this taken out event information and time information, from the start of statistical measurement to the present time Statistical information including the cumulative number for each event type and the sum of the time information from the start of statistical measurement to the present time is created and recorded in the statistical recording unit.

  Furthermore, in the third aspect of the present invention, in the apparatus according to the second aspect of the present invention, when the data amount determination unit is further (6) in the third stage, the usage rate is increased by the second. When the third rising threshold value, which is larger than the hour threshold value, is exceeded, it is determined that the amount of data has become the fourth stage which is larger than the third stage. (7) When the fourth stage is in the fourth stage, the usage rate is the second When the value falls below the third falling threshold value that is larger than the rising threshold value and smaller than the third rising threshold value, it is determined that the data amount has returned to the third stage. In the apparatus according to the second aspect of the present invention, when the data transfer unit further determines (5) the fourth stage, the event information and the time information are extracted from the internal buffer, and the next stage Nothing is transferred to and statistical information is created and recorded in the statistical recording unit.

  The outline of the present invention has been described above as an apparatus for controlling data transfer. However, in order to solve the above problems, the present invention further provides a method for performing the function by the apparatus. The present invention also provides a computer program for causing the apparatus to execute these methods.

  It should be noted that the above summary of the invention does not enumerate all necessary features of the present invention, and that combinations or subcombinations of these components can also be provided as the present invention.

  The best mode for carrying out the present invention (hereinafter referred to as an embodiment) will be described below in detail with reference to the accompanying drawings. Note that the same number is assigned to the same element throughout the description of the embodiment.

1. Configuration and Function of Apparatus According to the Implementation of the Present Invention The present invention is a measurement data processing apparatus 114 that is a main part of the “measurement apparatus” 104 in the measurement / recording / analysis environment of the real-time behavior of the observation target shown in FIG. Can be implemented as
FIG. 2 schematically shows the configuration and function of the measurement data processing apparatus 114 according to the embodiment of the present invention.

1.1 Storage of Measurement Data in Internal Buffer The data receiving unit 202 receives data obtained by sequentially measuring real-time events occurring in the observation target 102 as sequential event information. The data receiving unit 202 also generates time information related to the event information based on the absolute time when the sequential event information is received, which is measured by a clock or the like provided in the measurement data processing device 114.

Here, the time information can be an absolute elapsed time with the time when the measurement of the real-time event for the observation object 102 is started as 0.
However, since a large number of digits are generally required to represent the absolute elapsed time, it is inefficient from the viewpoint of the transfer bandwidth to transfer this as it is to the subsequent recording / analysis apparatus 106 as time information. Therefore, the time information can be a difference time from the time of the event measured immediately before to the time of the event measured this time. If the time when the measurement of the real-time event for the observation object 102 is started and the difference time between all the measurement events thereafter are known, the absolute elapsed time (that is, the occurrence time) of all the events can be known as a result. .

  For example, the data receiving unit 202 includes a differential time counter 204. When the differential time counter 204 receives the previous event information, the counter value is reset to start counting the time, and when the current event information is received. Counter value, that is, a difference time from the occurrence of the immediately preceding event to the occurrence of the current event can be generated as time information.

In this way, the sequentially received event information and the time information (for example, difference time) generated in association with the event information are sent together to the internal buffer 206 and temporarily stored therein.
FIG. 3A shows an example of the format of data that is sequentially sent from the data receiving unit to the internal buffer and stored therein. In this example, the time information is represented by an n-bit binary number. The time information that can be expressed in this format is 0 to (2 ⁿ ) -1 unit time.
In this way, a total of m + n bits of event information m bits and time information n bits are sequentially sent to the internal buffer 206 every time an event occurring in the observation target 102 is measured and received.

  The internal buffer 206 is a buffer (buffer) storage device for transferring measurement data and the like to the recording / analysis device 106 that is the next stage of the measurement data processing device 114, and is typically a FIFO (First In First). The buffer operates based on the principle of (Out). That is, the data stored in the buffer is discharged in order.

1.2 Determination of the level of the amount of data The data amount determination unit 208 monitors the usage rate of the data amount stored in the internal buffer 206 at that time in the total capacity of the internal buffer as needed, and It is determined to which of a plurality of stages the quantity level belongs.
FIG. 4 shows an example of the relationship between the usage rate of the monitored data amount in the total capacity of the internal buffer and the transition of the level of the data amount, which is determined by the data amount determination unit 208 based on the implementation of the present invention. Indicates. Hereinafter, description will be made along the example of this figure.

  First, in the initial state when the measurement data processing device 114 starts to operate, the data amount judgment unit 208 initially starts storing data gradually from a state in which no data is stored in the internal buffer. It is determined that the data amount is the lowest first stage (Low) 502. When the data amount is in the first stage and the usage rate of the internal buffer exceeds the first rising threshold value (for example, 45%), the data amount determination unit 208 has a larger data amount than the first stage. It is determined that the second stage (Middle) 504 is reached. Conversely, when the data amount is in the second stage and the internal buffer usage rate falls below a first falling threshold value (eg, 35%) that is smaller than the first rising threshold value, the data amount is in the first stage. It is determined that the state has returned to (Low).

  Further, when the data amount is in the second stage, the data amount determination unit 208 further determines the data when the internal buffer usage rate exceeds a second rising threshold value (for example, 75%) that is larger than the first rising threshold value. It is determined that the amount has reached the third stage (High) 506, which is larger than the second stage. Conversely, when the data amount is in the third stage, the internal buffer usage rate falls below a second falling threshold (eg, 65%) that is larger than the first rising threshold and smaller than the second rising threshold. Determines that the amount of data has returned to the second stage.

  Further, when the data amount is in the third stage, the data amount determination unit 208 further determines the data when the internal buffer usage rate exceeds a third rising threshold value (for example, 95%) that is larger than the second rising threshold value. It is determined that the amount has reached the fourth stage (Almost Full) 508, which is larger than the third stage. Conversely, when the data amount is in the fourth stage, the internal buffer usage rate falls below a third falling threshold value (for example, 85%) that is larger than the second rising threshold value and smaller than the third rising threshold value. Determines that the amount of data has returned to the third stage.

  In this way, by having a history (hysteresis) in the transition between the data volume stages, it is possible to prevent the determined data volume stages from being frequently switched when the internal buffer usage rate remains near the threshold boundary. can do.

  In the present embodiment, a mode in which the level of the data amount is four at the maximum is shown, but it is needless to say that a mode having more levels is naturally possible.

1.3 Change of Transfer Data Format The data transfer unit 210 sequentially extracts (ie, discharges) data including event information and time information for each event from the previously stored data from the internal buffer 206. Based on the level of the amount of data in the internal buffer 206 determined by the data amount determination unit 208, the format of the data is appropriately changed and transferred to the recording / analysis apparatus 106 as the next stage.
This format change is performed to avoid losing the measured data. In other words, as the amount of data to be transferred from the measuring device increases toward the limit of the transfer bandwidth, information that is indispensable for real-time measurement remains, but information that is difficult to adversely affect real-time measurement is omitted first. I will do it.

  If the level of the data amount determined by the data amount determination unit 208 is the first step (Low), the data transfer unit 210 does not change the format of the event information and the time information sequentially extracted from the internal buffer 206. Then, the data is transferred to the recording / analysis device 106 at the next stage as it is.

When the level of the amount of data is the second level (Middle), the data transfer unit 210 transfers the event information sequentially extracted from the internal buffer 206 without changing the format, and regarding the time information, This is converted into “rough time information” in which the accuracy of the time information is reduced to reduce the data amount, and the event information and the rough time information are transferred to the recording / analyzing device 106 in the next stage.
That is, when the level of the data amount is the second level (Middle), the amount of information is reduced with respect to the accuracy of time, but the permutation of measured events and a certain amount of time information between events can be obtained.

The coarse time information with reduced accuracy may be, for example, a reduced number of effective digits for expressing time information.
The coarse time information with reduced accuracy can be obtained by converting a numerical value representing the time information into a value of the number of digits of the numerical value. The number of digits means the leftmost digit when time information is expressed in binary. In other words, it is a value obtained by taking log2 of time information and rounding down the decimal point. For example, if the time information is 00000010, the coarse time information is 1, and if the time information is 00101001, the coarse time information is 5.
FIG. 3B shows an example of the format of data to be transferred when the level of the amount of data is the second level (Middle), and the case where the level is the first level (Low). )). Here, n> n ′.

Further, when the level of the amount of data is the third level (High), the data transfer unit 210 transfers the event information sequentially extracted from the internal buffer 206 without changing the format. The time information is not transferred directly to the next stage. Instead, statistical processing is performed on the data that has been extracted from the internal buffer so far, including the data extracted this time. Is recorded in the statistical recording unit 212 provided in the data transfer unit 210.
The result of the statistical processing recorded in the statistical recording unit 212 is transferred to the next-stage recording / analyzing device 106, either partly or entirely, periodically or under a predetermined condition, which will be described later. .
That is, when the level of the data amount is the third level (High), the permutation of the measured event is recorded in the subsequent stage, but the time information between events is not transferred every time, Transfer time information (ie, statistical information) in units of chunks.

Further, when the level of the data amount is the fourth level (Almost Full), the data transfer unit 210 does not transfer the event information and the time information sequentially extracted from the internal buffer 206 to the next level. As for the time information, as in the case of the third stage, the statistical processing is performed on the data that has been extracted from the internal buffer so far, including the data that has been extracted this time. The data is recorded in the statistics recording unit 212 provided in the data transfer unit 210.
A part or all of the results of the statistical processing recorded in the statistical recording unit 212 are transferred to the next-stage recording / analyzing device 106 periodically or at a predetermined condition.
That is, in the case of the fourth stage (Almost Full), the event permutation information is also omitted, but since the statistical processing is performed inside the apparatus, how much data the observation target was outputting, etc. It can be accurately captured.

1.4 Statistical Processing An example of statistical processing performed when the level of the amount of data is the third stage and the fourth stage will be described.
FIG. 5 shows an example of information managed by the statistical recording unit 212 in order to perform statistical processing.
First, when an event to be observed is measured, event information is included as an item in order to manage what the event is. The frequency is set to 0 at the start of statistical measurement, and each time the event is measured, the frequency is incremented by 1 and the frequency of the event is stored in the statistical recording unit 212. At the same time, the total elapsed time from the start of statistics measurement to the current time is obtained and recorded in the statistics recording unit 212. An average time interval can be calculated from the total elapsed time and frequency.
From these pieces of information recorded in the statistical recording unit 212, it is possible to know how often each event occurs in the real-time measurement and at what average time interval. By taking this information, it is possible to analyze how much measurement performance should be provided as hardware when actually analyzing in detail in real time.

The statistical recording unit 212 further includes at least one of the maximum value of the difference time for each event type from the start of statistics measurement to the current time, the minimum value, and the sum of squares of the difference time. be able to. This square sum value can be used to calculate the variance.
By always obtaining the mean and the variance, it is possible to obtain how much the data measured at a certain time is different from the data group measured so far. The deviation from the data group measured in the past may be treated specially because the measurement data may be some characteristic data. For example, not only the event information but also the time information may be transferred only for the observation data that is more than 3σ (σ is the standard deviation).

These pieces of information are updated every time measurement data comes in.
When event information that has not yet been entered in the statistical recording unit 212 is measured, the entry is newly added.
When all the entries in the event information column of the statistical recording unit 212 are filled and an event that does not exist in the entry is further measured, one entry is extracted from the statistical recording unit 212 and the extracted information is transferred to the recording / analyzing device at the subsequent stage To do. Which entry is to be extracted can be selected from the oldest updated entry based on the LRU (Least Recently Used) algorithm conventionally used in the cache technology, or can be selected by other methods. it can.

The event information measured most recently may be managed so that it moves to the top of the statistical recording unit 212 and is rearranged in order from the latest measured. Thus, the above LRU algorithm can be realized simply by managing the arrangement of tables. That is, events that are not measured much move down the table and eventually overflow from the table. The LRU algorithm can be implemented by transferring only the overflowed statistical information to the recording / analyzing apparatus in the next stage. As a result, events that occur frequently are statistically processed inside the measurement data processing device and are not transferred to the recording / analysis device at the subsequent stage. Go down.
When the usage rate of the internal buffer decreases, all or part of the statistical information recorded in the statistical recording unit 212 can be transferred to a subsequent apparatus to clear the statistical recording unit 212. For example, on the condition that the data amount changes from the third stage (High) or the fourth stage (Almost Full) to the second stage (Middle), the statistical information of the statistical recording unit 212 is transferred to the subsequent stage and the statistical recording. The section 212 can be cleared.
The clearing of the statistical recording unit 212 can also be performed based on temporal conditions. For example, until the predetermined time elapses from the time when this clearing was performed last time, even if the data amount is the second stage (Middle), the statistical information in the statistical recording unit 212 is transferred to the subsequent stage and the statistical It is possible to prevent the recording unit 212 from being cleared.

  As described above, even when the amount of measured data is large, it is possible to obtain useful information such as statistical information while avoiding the situation of losing information represented by the measurement data.

1.5 Time Adjustment Unit If the method according to the embodiment of the present invention described so far is used, the accuracy of time information is reduced when the amount of data is medium or large. Therefore, the absolute time from the start of measurement can be periodically output to the subsequent apparatus. As a result, the absolute time from the start of measurement can be corrected in a timely manner, and the absolute time can be recovered with the measurement start time set to 0 even using this method for changing the format of the transfer data. Hereinafter, this embodiment will be described.

The measurement data processing device 114 includes a time adjustment unit 222. The time adjustment unit 222 includes an absolute time counter 224 that manages an absolute time in which the measurement start time of the observation target is zero.
FIG. 6 shows an example of an absolute time format counted by the absolute time counter 224.
The absolute time is required when the event occurrence frequency is extremely low, that is, when the time information of the differential time counted by the differential time counter 204 overflows. Therefore, when the difference time counter overflows, the absolute time of the absolute time counter 224 is sent to the subsequent apparatus, and at the same time, the difference time counter is reset.

  FIG. 7 shows an example of the absolute time transmission timing by the time adjustment unit 222. When there is no event information to be measured, the absolute time T1 is sent when the difference time from the measurement of the immediately preceding event information overflows the difference time counter. After that, with the differential time counter reset, it waits for event information to be measured again. When the event information is not yet measured and the difference time again overflows the difference time counter, the absolute time T2 is sent and the difference time counter is reset. Thereafter, when a certain amount of time elapses from T2 and event information is measured, the event information and time information of a difference time from T2 are transmitted.

2. As described above, the embodiment of the present invention has been described as a measurement data processing apparatus that measures an event occurring in an observation target and transfers the measured data to the next stage. It can also be understood from the aspect as a method of controlling data transfer by the measurement data processing apparatus.

FIG. 8 is a diagram showing a flow of a processing procedure by the measurement data processing apparatus when the present invention is viewed from the aspect as a method for controlling data transfer.
It should be noted that the values in this figure are merely examples of implementation and can be changed to other values.

First, in step 804, the previous determination result of the data amount of the data stored in the internal buffer provided in the measurement data processing device is examined.
If the data amount determined last time is the first stage (Low), the current data amount is checked in step 804. If the ratio of the current data amount to the internal buffer is 45% or more, the current data amount is The amount is determined to be the second step (Middle), and if it is less than 45%, it is determined that the current data amount remains the first step (Low).
Similarly, if the previously determined data amount is the second stage (Middle), the third stage (High), and the fourth stage (Almost Full), respectively, Step 806, Step In step 808 and step 810, the current amount of data is checked, and in step 814, step 816, step 818, and step 820, the current amount of data is the first stage ( Low), second stage (Middle), third stage (High), and fourth stage (Almost Full).

In step 822, it is checked whether or not the judgment result of the current data amount has changed from the third stage (High) to the second stage (Middle). If so, in step 824, the statistical data provided in the measurement data processing apparatus is provided. All the statistical information recorded in the recording unit is sent to the next stage, the statistical recording unit is cleared, and the process proceeds to step 826.
If it has not changed from the third stage (High) to the second stage (Middle), a pair of event information and time information are read from the internal buffer in Step 826.

Next, in step 828, the data amount determined earlier is checked, and if it is the first stage (Low), event information and time information read from the internal buffer are transferred to the subsequent apparatus in step 832. .
If the data amount is in the second stage (Middle), in Step 834, the event information read from the internal buffer and the coarse time information with reduced accuracy of the time information are transferred to the subsequent apparatus.
If the data amount is the third stage (High), in Step 836, only the event information read from the internal buffer is transferred to the subsequent apparatus.
When the data amount is the fourth stage (Almost High), neither event information nor time information is transferred to the subsequent apparatus.
If the amount of data is the third stage (High) or the fourth stage (Almost High), it is further determined in step 838 whether the event information is recorded in the statistical information entry. In step 840, the event information is added to the statistical information as a new entry. Thereafter, in step 842, the frequency field related to the event information is increased by 1, and the value of the total elapsed time is updated.

  Thereafter, in step 844, it is determined whether or not a predetermined sufficient time has passed since the previous transmission of the statistical information. If it has elapsed, in step 846, all the recorded statistical information is transmitted to the next stage. However, this statistical recording section is cleared.

As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range of description of the said embodiment.
It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. Further, it is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention. Furthermore, it should be noted that not all the combinations of features described in the above embodiments are essential for the solution of the invention.

An example of the measurement, recording, and analysis environment for the real-time behavior of the observation target is shown. The structure of the measurement data processing apparatus based on implementation of this invention and its function are shown typically. An example of a format of data stored in an internal buffer and an example of a format of data in which the accuracy of time information transferred to the subsequent stage is lowered in comparison with the data are shown. The example of the relationship between the usage rate which occupies for the internal buffer of the data amount judged based on implementation of this invention, and the transition of the grade of the amount of data is shown. The example of the information managed in order to perform a statistical process is shown. The example of the format of the absolute time counted by the absolute time counter is shown. The example of the transmission timing of the absolute time by a time adjustment part is shown. The flow of a processing procedure when the present invention is viewed from the aspect as a method for controlling data transfer will be described.

Claims (16)

A measurement data processing device that measures an event occurring in an observation target and transfers the measured data to the next stage.
A data receiving unit that receives event information indicating a type of the event that has occurred, and generates time information related to the event information including an absolute elapsed time from the start of the measurement to the occurrence of the event When,
An internal buffer for temporarily storing the event information and the time information;
A data amount determination unit for determining the amount of data stored in the internal buffer;
(1) When the data amount is in the lowest first stage, and the usage rate of the data amount in the internal buffer exceeds a first rising threshold, the data amount is less than the first stage. Judging that it was the second stage,
(2) In the second stage, when the usage rate falls below a first falling threshold value that is smaller than the first rising threshold value, it is determined that the data amount has returned to the first stage;
A data amount determination unit;
If it is determined that it is the first stage, the event information and the time information are extracted from the internal buffer, and the extracted event information and the time information are transferred to the next stage, When determining that it is the second stage, the event information and the time information are extracted from the internal buffer, and the event information and the rough time information in which the accuracy of the time information is reduced, A data transfer unit to transfer to the next stage;
A data measuring device comprising: The data amount determination unit further includes:
In the second stage, if the usage rate exceeds a second rising threshold value that is greater than the first rising threshold value, the data amount has become a third stage that is greater than the second stage. Judgment
In the third stage, if the usage rate falls below a second falling threshold value that is greater than the first rising threshold value and smaller than the second rising threshold value, the data amount is reduced to the second stage. Judge that he has returned,
The data transfer unit further includes:
Including the statistics recording section,
If it is determined that it is the third stage, the event information and the time information are extracted from the internal buffer, and only the extracted event information is transferred to the next stage. Statistical information generated based on the event information and the time information, including the cumulative number of each event type from the start of statistical measurement to the current time, and the total elapsed time from the start of statistical measurement to the current time , Create statistical information and record it in the statistical recording section,
The apparatus of claim 1. The data amount determination unit further includes:
In the third stage, if the usage rate exceeds a third rising threshold value that is greater than the second rising threshold value, the data amount is in a fourth stage that is greater than the third stage. Judgment
In the fourth stage, if the usage rate falls below a third falling threshold value that is greater than the second rising threshold value and smaller than the third rising threshold value, the data amount is reduced to the third stage. Judge that he has returned,
The data transfer unit further takes out the event information and the time information from the internal buffer when determining that it is the fourth stage, transfers nothing to the next stage, and Create the statistical information and record it in the statistical recording unit,
The apparatus of claim 2. The data receiving unit includes a differential time counter,
The time information includes a difference time from a last received event to a currently received event measured by the difference time counter.
The apparatus according to claim 2 or 3.   The apparatus according to claim 4, wherein the coarse time information includes a format in which the number of significant digits representing the time information is reduced.   5. The apparatus according to claim 4, wherein the coarse time information includes a numerical value representing the time information converted into a value of the number of digits of the numerical value.   The statistical information further includes a maximum value of the difference time for each event type from the start of statistical measurement to the current time, a minimum value of the differential time for each event type from the start of statistical measurement to the current time, The apparatus according to claim 4, comprising at least one of a sum of squares of the difference time from the start of measurement to the present time.   The data transfer unit further discharges a part of the statistical information recorded in the statistical recording unit and transfers it to the next stage when the statistical recording unit becomes full. Equipment.   The discharge of a part of the statistical information recorded in the statistical recording unit is based on an LRU (Least Recently Used) algorithm, and the most recently updated event type among the information for each event type The apparatus according to claim 8, comprising discharging information related to.   The data transfer unit further includes all or one of the statistical information recorded in the statistical recording unit in response to the data amount being changed from the third stage or the fourth stage to the second stage. The apparatus according to claim 4, wherein a part is discharged and transferred to the next stage.   The data transfer unit is further responsive to the elapse of a predetermined time from the time when all or part of the statistical information recorded in the statistical recording unit was previously discharged and transferred to the next stage, The apparatus according to claim 4, wherein all or part of the statistical information recorded in the statistical recording unit is discharged and transferred to the next stage. A time information recovery unit comprising an absolute time counter for managing the absolute time from the measurement start time of the observation target;
When the differential time counter overflows, the time information recovery unit transfers the absolute time in that case to the next stage, and resets the differential time counter.
The apparatus according to claim 4. A method of processing measurement data in a device that measures an event occurring in an observation target and transfers the measured data to the next stage,
Receiving event information indicating a type of the event that has occurred, and generating time information related to the event information including an absolute elapsed time from the start of the measurement to the occurrence of the event;
Temporarily storing the event information and the time information in an internal buffer in the device;
Determining the amount of data stored in the internal buffer,
(1) When the data amount is in the lowest first stage, and the usage rate of the data amount in the internal buffer exceeds a first rising threshold, the data amount is less than the first stage. Judging that it was the second stage,
(2) In the second stage, when the usage rate falls below a first falling threshold value that is smaller than the first rising threshold value, it is determined that the data amount has returned to the first stage;
Steps,
If it is determined that it is the first stage, the event information and the time information are extracted from the internal buffer, and the extracted event information and the time information are transferred to the next stage, When determining that it is the second stage, the event information and the time information are extracted from the internal buffer, and the event information and the coarse time information in which the accuracy of the time information is reduced, Transferring to the next stage;
A method comprising: The step of determining further comprises:
In the second stage, if the usage rate exceeds a second rising threshold value that is greater than the first rising threshold value, the data amount has become a third stage that is greater than the second stage. Judgment
In the third stage, if the usage rate falls below a second falling threshold value that is greater than the first rising threshold value and smaller than the second rising threshold value, the data amount is reduced to the second stage. Judge that he has returned,
If it is determined that the transferring step is the third stage, the event information and the time information are extracted from the internal buffer, and only the extracted event information is transferred to the next stage. The statistical information generated based on the event information and the time information transferred and taken out, and the cumulative number for each type of event from the start of the statistical measurement to the current time, and the current time from the start of the statistical measurement Including the sum of the time information up to and creating statistical information and recording it in a statistical recording unit in the data measuring device,
The method of claim 13. The step of determining further comprises:
In the third stage, if the usage rate exceeds a third rising threshold value that is greater than the second rising threshold value, the data amount is in a fourth stage that is greater than the third stage. Judgment
In the fourth stage, if the usage rate falls below a third falling threshold value that is greater than the second rising threshold value and smaller than the third rising threshold value, the data amount is reduced to the third stage. Judge that he has returned,
If it is determined that the transferring step is the fourth stage, the event information and the time information are extracted from the internal buffer, nothing is transferred to the next stage, and Create the statistical information and record it in the statistical recording unit,
The method according to claim 14. A computer program for controlling data transfer in a device that measures an event that occurs in an observation target and transfers the measured data to the next stage.
Receiving event information indicating a type of the event that has occurred, and generating time information related to the event information including an absolute elapsed time from the start of the measurement to the occurrence of the event;
Temporarily storing the event information and the time information in an internal buffer in the device;
Determining the amount of data stored in the internal buffer,
(1) When the data amount is in the lowest first stage, and the usage rate of the data amount in the internal buffer exceeds a first rising threshold, the data amount is less than the first stage. Judging that it was the second stage,
(2) In the second stage, when the usage rate falls below a first falling threshold value that is smaller than the first rising threshold value, it is determined that the data amount has returned to the first stage;
Steps,
If it is determined that it is the first stage, the event information and the time information are extracted from the internal buffer, and the extracted event information and the time information are transferred to the next stage, When determining that it is the second stage, the event information and the time information are extracted from the internal buffer, and the event information and the coarse time information in which the accuracy of the time information is reduced, Transferring to the next stage;
A computer program that runs

Images