JP6524848B2 - Failure prediction device, failure prediction method and failure prediction program - Google Patents

Description

  The present invention relates to a disk drive failure prediction technique.

  In the hard disk, a defective sector may be generated on the disk surface due to a minute defect and reading and writing can not be performed, or the data transfer speed may be significantly reduced by repeating the retry operation. In addition, there is also a possibility that a failure may expand and cause the hard disk itself not to start up.

  Patent Document 1 discloses a technique of measuring a transfer time of a hard disk and detecting a sign of failure of the hard disk from the transfer time. When the transfer time is measured every predetermined time (for example, one week) after product shipment and compared with the factory transfer time, the difference between the two transfer times satisfies the predetermined condition, It performs warning notification processing that there may be a problem with the hard disk in the future.

JP, 2011-68109, A

  In the prior art disclosed in Patent Document 1, a value obtained by multiplying the measurement value of transfer time by a predetermined coefficient is compared with the measurement time at the time of factory shipment, and when the former is long, it is determined to be a delay region. Then, when the delay area is equal to or more than a predetermined threshold value, a warning is issued. However, the variation in access time depending on the head position of the hard disk drive (HDD) has not been taken into consideration at all. For example, even if the same area (sector) of HDD is accessed, if the rotational latency time is maximum, the latency time for one round of the disk will be required, and the access time will be longer by that time, and the rotational latency time will be In the case of the minimum, access time is shortened because access can be made at the shortest without waiting time for disk rotation. As described above, even when the same area of the HDD is accessed, the access time until the completion of data reading fluctuates depending on the head position before access and the timing of access, but in the prior art, such fluctuation is The access time could not be measured with sufficient accuracy because it was not taken into consideration. Therefore, even if failure prediction is performed based on such access time, failure prediction of the HDD can not be performed with high accuracy.

  The present invention has been made in view of these circumstances, and an object thereof is to provide a failure prediction technique capable of predicting the occurrence of a failure in a disk drive using device and preventing the loss of data in the disk. is there.

  In order to solve the above problems, a failure prediction apparatus according to an aspect of the present invention includes an abnormal value recording unit that stores an upper limit threshold of access time when an access target location is accessed according to a predetermined access pattern for a disk drive; And a control unit that registers the access target location in the abnormal value recording unit when the measured value of the access time when the access target location is accessed according to the access pattern exceeds the upper limit threshold.

  Another aspect of the present invention is a failure prediction method. This method determines the upper threshold of the access time when the access target location is accessed according to a predetermined access pattern to the disk drive and records the upper threshold in the abnormal value recording unit, and accesses the access target location according to the access pattern And registering the location to be accessed in the abnormal value recording unit, when the measured value of the access time at the time of execution exceeds the upper threshold value.

  Note that arbitrary combinations of the above-described components, and conversions of the expression of the present invention among methods, apparatuses, systems, recording media, computer programs and the like are also effective as aspects of the present invention.

  According to the present invention, it is possible to predict the occurrence of a failure in a disk drive using device and to prevent the loss of data in the disk.

FIG. 1 is a configuration diagram of an HDD failure prediction device according to an embodiment. It is a flowchart which shows the failure prediction procedure by the HDD failure prediction apparatus of FIG. It is a flowchart which shows the detailed procedure of the failure prediction process of FIG. It is a flowchart which shows the detailed procedure of the warning determination processing of FIG. It is a flowchart which shows the detailed procedure of the error determination processing of FIG. It is a figure explaining the access pattern for measuring command execution time. FIG. 7 is a diagram for explaining command execution time measured by the access pattern of FIG. 6; It is the graph which showed the schematic diagram of FIG. 7A by actual value. It is a figure explaining evaluation data and threshold data in a segment which grouped command issue numbers. FIG. 9A shows the access time of the HDD 300 in the normal state, and FIG. 9B shows the access time of the HDD 300 in the abnormal state. FIG. 10A shows an abnormal value database in which command issue numbers and threshold values are associated with each other and FIG. 10B is a diagram showing an abnormal value database in which segment numbers and threshold values are associated and recorded. is there. FIG. 11A shows a warning counter counted for each command issue number, and FIG. 11B shows an error counter counted for each command issue number.

  If there is a problem with the operation of the HDD, there is no guarantee of the written data. Once the operation failure of the HDD occurs, the internal data can not basically be read out, resulting in a large loss. That is, at the stage of noticing that it is a failure, it can not be avoided to lose data anywhere in the HDD. Therefore, in the HDD failure prediction apparatus 100 according to the embodiment of the present invention, the lifetime of the HDD is predicted as the limit at which the normal reading of data can be made normal, and the loss of data due to the failure of the HDD is avoided. With the goal.

  Patent Document 1 describes a method of evaluating an access time for predicting the lifetime of an HDD by comparing transfer times. However, with regard to HDD failures, in particular the failures resulting from the wear of internal parts due to age-related deterioration tend to expand with time, not just at the position where the failure currently occurs, so we do not capture the sign with high accuracy And, with the passage of time, the possibility of losing internal data on the HDD increases.

  In particular, since re-accessing an area progressing to a failure is likely to progress to a serious failure, being able to predict the progress of the failure determines the method for taking measures for the subsequent HDD internal data recovery. It will be a very important guide above.

  Therefore, in HDD failure prediction apparatus 100 of the present embodiment, the accuracy of failure prediction is determined by determining the threshold used when predicting failure based on access time based on the scatter diagram of access time based on a specific access pattern. Adopt a method to raise

  FIG. 1 is a block diagram of an HDD failure prediction apparatus 100 according to the embodiment. The HDD failure prediction apparatus 100 has a function of driving the HDD 300 used by the host 200 and a function of predicting failure of the HDD 300. The HDD failure prediction apparatus 100 includes an HDD controller 10, a temporary storage unit 20, a control unit 30, and an abnormal value DB recording unit 40. These configurations can be realized by hardware, software, or a combination thereof. The HDD failure prediction apparatus 100 and the host 200 can be integrally configured. The HDD failure prediction apparatus 100, the host 200, and the HDD 300 may be integrally configured.

  The HDD controller 10 receives a read / write command for the HDD 300 from the host 200 based on the ATA standard of the hard disk drive, writes data to the HDD 300, and reads data from the HDD 300. Temporary storage unit 20 temporarily stores the transfer data in a FIFO structure.

  Control unit 30 measures a command execution time (access time) at the time of writing of HDD controller 10 and determines the life of HDD 300.

  The abnormal value DB recording unit 40 stores a threshold value for determining an abnormality in access time, and registers, as a database, a command issue number and a delayed access time at which the delayed access occurred when the delayed access occurred. In the present embodiment, as described later, the HDD 300 is accessed in a predetermined number of sector units (specific capacity units). In correspondence with the first LBA (Logical Block Address) used for each access, 1, 2, 3,. . . The one given a serial number such as is called a command issue number. In the following description, it is also possible to use an LBA instead of the command issue number. However, using the command issue number can reduce the data capacity required for storage and operation.

  The host 200 includes a display unit and an input unit (not shown). The input to the input unit of the host 200 is transmitted to the control unit 30 of the HDD failure prediction apparatus 100 and processed. The control unit 30 of the HDD failure prediction apparatus 100 also functions as a display control unit that controls the display unit of the host 200.

  When the host 200 issues a data write command to the HDD 300, the data to be written is temporarily stored in the temporary storage unit 20 via the HDD controller 10. When the temporarily stored write data reaches a predetermined capacity, the HDD controller 10 writes on the HDD 300 according to the access pattern shown in FIG. Details of this process will be described later.

  At this time, the control unit 30 measures a command execution time (access time) when data is written from the HDD controller 10 to the HDD 300, reads a threshold value stored in advance in the abnormal value DB storage unit, and obtains the current access time. Determines whether the threshold exceeds this threshold.

  When it is confirmed that the access time exceeds the threshold, the control unit 30 stores the LBA where the access time exceeds the threshold and the access time, counts the warning (warning) counter and the error counter described later, and gives a warning Failure of the HDD 300 is predicted by transition of the counter and the error counter, and when it is predicted that the life has been reached, a stop warning of the HDD is issued.

  Here, an overview of the access time of the HDD 300 will be described. FIG. 9A shows the access time of the HDD 300 in the normal state, and FIG. 9B shows the access time of the HDD 300 in the abnormal state.

  As shown in FIG. 9A, the access time in the normal state can be represented by the sum of the time dependent on the head operation such as the seek time, rotational latency time, focusing time, etc., and the data transfer time.

  As shown in FIG. 9 (b), the access time at the time of abnormality occurrence is several tens of times longer at the time of normal access since the retry time and the alternative sector processing time etc. are added at the time of the alternative sector in addition to the access time at the normal time. Processing time is required.

  In the HDD 300, since the disk is rotating at a constant speed, when the head accesses the target sector, depending on the access timing, a rotational waiting time corresponding to one disk rotation occurs at the maximum. Further, since the seek time to the position to be accessed is fluctuated depending on the head position when the previous access is completed, the total access time is also fluctuated. As a result, since the convergence time until access is also different, the target sector can not be accessed in the same access time as factory shipment, and it varies from access to access.

  The failure determination method of Patent Document 1 has a problem that it is not possible to predict failure of the HDD with high accuracy because both the data at the time of factory shipment and the data at the time of life evaluation have variations. Therefore, in the present embodiment, the variation of the access time is dealt with by using the access pattern shown in FIG.

  FIG. 2 is a flowchart showing a failure prediction procedure by the HDD failure prediction apparatus 100.

  In step S201, the HDD controller 10 accesses the HDD 300 according to the access pattern of FIG. 6, and the control unit 30 measures the command execution time at that time.

  In the access pattern shown in FIG. 6, in the present embodiment in which data of a specific capacity is written corresponding to one command issue number, the specific capacity is 256 sectors, but the number of sectors other than that is However, the present invention is not limited to this. In the case of the HDD 300, if the head position at the end of the previous access is not specified, the seek time particularly fluctuates, and a relatively accurate access time can not be measured. Therefore, as shown in FIG. 6, after the head position has been accessed, it is always reset to the initial position (here, the position of sector 0) and then the specific capacity (256 sectors) is sequentially written to achieve more accurate access time. It is possible to measure the

  FIG. 7A is a schematic diagram of access time when the HDD 300 is accessed according to the access pattern of FIG. The horizontal axis is the command issue number, and the vertical axis is the access time for each command issue number. FIG. 7B is a graph showing the schematic view of FIG. 7A by actual measurement values. When the measurement data is plotted, a scatter plot as shown in FIG. 7B is obtained, and the access times are distributed in a band with rising right. Hereinafter, a disk area (sector) of a specific capacity accessed for each command issue number will be referred to as a “command issue number area”.

  In an ideal state in which the HDD 300 has no rotational delay and no convergence time, a scatter plot obtained by plotting the access time to the sector serving as the access target should be approximately one line. However, as described above, since there is always variation in rotational latency and focusing time for disk access, when the access time for each specific capacity shown by the access pattern in FIG. 6 is plotted, it is actually identified as in FIG. 7B. There is a band-like shape with a variation of, and there is a strong correlation between the command issue number and the access time. The inclination of the entire band in FIG. 7B is mainly due to the increase in seek time due to the sector to be accessed becoming far from the reset position of the head as the command issue number increases as shown in FIG.

  The command execution time (access time) may be measured by the HDD failure prediction apparatus 100 or by an external device that generates the same access pattern.

  Next, in step S202, a value corresponding to the upper end of the band in the scatter diagram measured in step S201 is detected. Then, this value is used as a threshold (failure prediction threshold) for each command issue number. This value is the access time when the waiting time depending on the head operation such as the seek time, the rotational waiting time, and the focusing time is maximal, and when the HDD 300 is normal, the access time falls below this value It is characterized by Therefore, by using this value as a failure prediction threshold, it can be understood that the access of the HDD 300 is not normal if the access time exceeds this threshold. A specific method of detecting this threshold will be described later.

  Next, in step S203, the threshold value calculated in step S202 is registered in the abnormal value DB recording unit 40. Specifically, as shown in FIG. 10A, the command issue number and the threshold value are associated and recorded.

  Steps S201 to S203 are pre-processing or initial setting processing for performing failure prediction.

  Next, in step S204, a failure prediction operation is performed when using the HDD 300. The control unit 30 monitors whether the access time of the command issuance area exceeds the threshold recorded in the abnormal value DB recording unit 40, and the command issuance number in the abnormal value DB recording unit 40 for the command issuance number area exceeding the threshold And record the access time. The control unit 30 counts the warning counter and the error counter recorded in the abnormal value DB recording unit 40, and as a result, when it reaches a level determined to be a failure prediction, it issues a stop warning of the HDD 300 and terminates the process. Do.

  In the processes of steps S201 to S203, the control unit 30 operates as a threshold calculation unit. The threshold calculation unit may be a circuit different from the control unit 30.

  If the HDD model number and the firmware are the same, the threshold obtained from the band of the scatter diagram in FIG. 7 is the same, and there is no need to create a new threshold, so steps S201 to S203 are omitted. The failure prediction in step S204 can be started using the threshold measured by another HDD. When using the threshold measured with another HDD, it is not necessary to provide the HDD failure prediction apparatus 100 with a threshold calculation unit.

  Here, the method of detecting the threshold in step S202 will be described in detail. The first method of step S202 will be described.

  As shown in FIG. 8, command issue numbers are grouped for each of a predetermined number N of command issue numbers. In the following, this group is referred to as "command issuing segment" or simply "segment". Also, the predetermined number N is referred to as "segment length". Here, the predetermined number N (segment length) is set such that one segment includes approximately one or more points at which the waiting time depending on the head operation such as the seek time, rotation waiting time, focusing time, etc. becomes maximum. Do. Typically, N = 30 to 50. For example, when N = 30, the command issue number = 1 to 30 is segment 1, command issue number = 31 to 60 is segment 2, command issue number = 61 to 90 is segment 3, and so on. Corresponds to the segment.

  Next, the maximum value of access time is detected for each segment. And let the maximum value be a threshold in the segment. For example, when N = 30 and the access time is maximum and the maximum value is 30 msec in command issue number = 12 in segment 1, all the threshold values corresponding to command issue numbers 1 to 30 are 30 msec. . Alternatively, a value obtained by multiplying the maximum value of the access time in each segment by a predetermined multiplying factor may be used as the threshold value of the segment. For example, the predetermined magnification may be 1.2, and the maximum value 30 msec × 1.2 = 36 msec may be used as the threshold value of the segment. Alternatively, the threshold may be a value obtained by adding a predetermined value to the maximum value of the access time in each segment. For example, the predetermined value may be 5 msec, and the maximum value 30 msec + 5 msec = 35 msec may be a threshold.

  The threshold values calculated by the first method all have the same value in the command issue number corresponding to one segment. Therefore, instead of recording the threshold for each command issue number in step S203, as shown in FIG. 10B, the segment number may be recorded in correspondence with the threshold.

  The second method of step S202 will be described. First, as in the first method, a segment is formed for each predetermined number of command issue numbers. This segment is for use in the subsequent processing steps, and in step S202, the segment is not used.

  Next, for a certain command issue number (command issue number i), the maximum value of the access time is detected for command issue numbers (i-w to i + w) in a predetermined range before and after that. That is, the threshold value θ [i] corresponding to the command issue number i is calculated according to the equation (1). Here, a [i] is the command execution time (access time) corresponding to the command issue number i, w is a positive integer, and max is a function that returns the maximum value among the values specified in the arguments. is there. According to Equation (1), the maximum value is detected for (2w + 1) command issue numbers. The positive integer w may be set such that one or more local maximum values of the access time are included in the (2w + 1) command issue numbers. Typically, w = 15 to 25 may be used. (2w + 1) may be the same as or different from the segment length N. Assuming that the maximum command issuance number used for measurement in step S201 is P, threshold values θ [w + 1] to θ [Pw] corresponding to i = (w + 1) to (Pw) are respectively calculated according to the equation . For i = 1 to w, θ [w + 1] may be diverted, and for i = (P−w + 1) to P, θ [P−w] may be diverted.

  Further, the threshold may be calculated by further performing moving average processing on the value calculated according to the equation (1). For example, the left side of equation (1) is substituted into the temporary variable μ [i], and the moving average of μ [i] is calculated according to equation (2) to obtain the threshold θ [i]. Here, ε [j] is a weighting factor that satisfies the equation (3). In addition, L is a positive integer, and may be typically set to 5 to 10. By calculating the threshold value in accordance with Equation (2), the change in the threshold value may be smooth, and failure may be predicted accurately.

  FIG. 3 is a flowchart showing the detailed procedure of the failure prediction process shown in step S204.

  The failure of the HDD 300 is such that the access time of the area divided by the specific capacity unit exceeds the threshold which is the upper end of the band of the scatter diagram shown in FIG. It predicts depending on whether it spreads again.

  The cause of the access delay exceeding the threshold value of the HDD 300 is due to the addition of processing time at the time of abnormality occurrence such as retry and alternative sector generation in addition to processing at the time of normal access since abnormality occurred at the time of access.

  However, although the occurrence of a substitute sector due to the damage of only a specific sector causes a large access delay only when the substitute sector occurs, the HDD 300 is regarded as returning to normal operation even if the same sector is accessed thereafter. There is a feature that access delay does not occur even if the same area is accessed again. In this case, the occurrence is also single and the expansion of the abnormal sector can not be confirmed, and the expansion beyond the alternative sector generated at that time can not be seen. On the other hand, as the access delay due to age deterioration progresses, the number of areas exceeding the threshold of the area divided by the segment length gradually increases with each access, but since the pattern can not be identified, it is separated by segments Failure is predicted by registering the transition of how the threshold value of the area is exceeded in the abnormal value DB recording unit 40.

  In step S301 in FIG. 3, it is specified to which command issue number the top data transferred from the temporary storage unit 20 to the HDD 300 corresponds. For example, in temporary storage unit 20, data to be transferred is managed using the same LBA as that normally written from host 200 to HDD 300, and a value obtained by dividing the head LBA by the specific capacity at the time of transfer is calculated do it. The specified command issue number (the first command issue number used for data transfer) is i.

  In step S305, it is calculated how many command issuance numbers (specific capacities) correspond to the data capacity (write capacity) transferred from the temporary storage unit 20 to the HDD 300. Specifically, the data capacity is divided by the specific capacity, and the quotient and the remainder are calculated. And let the quotient be the number of times of writing M.

  Writing to the HDD 300 can be controlled in units of sectors, but since the specific capacity is larger (here, 256 sectors), when the last write data is less than the specific capacity, existing data is within the specific capacity to be written last It may exist.

  Therefore, in step S310, it is determined whether the remainder calculated in step S305 is "0". That is, it is determined whether the write capacity is divisible by the specific capacity. As a result, if division is not possible (if there is a remainder) (NO in S310), the process proceeds to step S315. In step S315, it is checked whether or not there is other data in the M + 1th write area corresponding to the fractional data, and if there is other data (YES in S315), the existing data is temporarily stored in step S320. After the data is read into the storage unit 20 and combined with the (M + 1) th write data, it is prepared as the (M + 1) th write data, and the process proceeds to step S325. If there is no other data in step S315 (NO in S315), it is not necessary to combine the write data, so the process directly proceeds to step S325. As a result, since the number of times of writing increases by one for fractional minutes, M is added by 1 in step S325, and the process proceeds to step S330.

  If the write capacity is divisible by the specific capacity in step S310 (YES in S310), the process proceeds to step S330.

  In step S330, the specific capacity corresponding to the command issue number i in the access pattern shown in FIG. 6 is written.

  In step S350, a warning determination process for detecting a non-fatal failure is performed. The warning determination process will be described in detail later with reference to the flowchart of FIG.

  In step S355, an error determination process is performed to detect a fatal failure. In this error judgment processing, the end of the life is detected by monitoring a fatal abnormal sector which is not expected to be recovered, and at the time of the detection of the end of the life, an operation stoppage of the HDD 300 is finally issued. To evacuate data from the HDD 300 of FIG. The error determination process will be described in detail later with reference to the flowchart of FIG.

  In step S360, the warning counter value and the error counter value of HDD 300 updated in the process of steps S350 and S355 are displayed. The user (operator) can monitor the state of the HDD 300 based on these counter values, and when necessary, the operator can judge from this numerical value and stop the HDD 300 independently.

  In step S365, when it is confirmed that the error counter processed in response to the result of step S360 exceeds the operation stop parameter value (YES in S365), the failure prediction process is ended. A warning message may be displayed immediately after “YES” in step S365, and a message may be displayed to notify the user of the failure prediction process.

  While data is written to the HDD 300, the processes of steps S350 and S355 are performed on the command issue number to predict failure of the HDD 300 being used. Therefore, the number M of times of writing is decremented by one in step S370, and the command issuance area i is incremented by one because the access is shifted to the next area.

  Finally, in step S375, when the number M of times of writing is greater than 0 (YES in S375), the series of processes from step S330 to step S370 is repeated until the predetermined number of times of writing is reached. When the number M of times of writing becomes 0 (NO in S 375), the failure prediction process is ended.

  FIG. 4 is a flowchart showing a detailed procedure of the warning determination process of step S350. In the warning determination process, an area in which an abnormality has occurred is identified, and it is determined how the abnormality occurrence area is distributed (is spread).

  The warning determination is performed by adding the value of the warning counter for each command issue number shown in FIG.

  In step S401, it is determined whether the command execution time a [i] corresponding to the command issue number i has exceeded the threshold value θ [i].

  As a feature of the access time of the HDD 300, since the access is completed within the normal processing time, the access time of the segment falls within the threshold because the HDD 300 in which failure or deterioration has not progressed is completed within the normal processing time. However, in the HDD 300 which has deteriorated with age and the HDD 300 which has developed a fault, the internal fault is enlarged due to the contamination of the head, the influence of the dust accumulated inside, and the scratches generated on the board surface. Since the time requires processing corresponding to the fault and the internal processing takes more time than normal processing, the access time exceeds a predetermined threshold, and the delay of the access time to the write address is extended. In step S401, it is checked whether such a symptom has occurred.

  If the command execution time exceeds the threshold (YES in S401), the warning counter is increased by one in step S402, and the command issue number and the command execution time (access time) at that time are registered in the abnormal value DB recording unit 40. Do.

  If the command execution time does not exceed the threshold (NO in S401), the process proceeds to step S403. If the past warning corresponding to the command issue number is counted in step S403 (YES in S403), it is considered that there is a temporary factor at the time of access, and the warning count in the command issue number area is 0 in step S407. The warning determination process is ended.

  Next, the process proceeds from step S402 to step S404. In step S404, in order to reliably detect the spread of the fault, in the narrow neighboring segment area (first neighboring segment area) in which five segments before the segment to be accessed and five segments later are taken, as shown in FIG. Check if the scatter plot band is spreading beyond the threshold. Specifically, it is determined whether ten or more non-zero warning counters are generated in the first neighboring segment area. When ten or more warning counters are confirmed (YES in step S404), it is determined that the delay is increased in this neighboring segment area, and a warning of stop of the HDD 300 can be issued in the next error determination processing, In step S405, a threshold value (10 in this case) of failure prediction is added to the error counter.

  For the damage caused by the contact of the head physically generated on the disk, the continuity of the sector can be seen in the access delay, and the cause of the occurrence can be identified relatively easily, but in the life prediction, the access delay occurs once. Since the access delay is characterized by the occurrence of the access delay in its neighboring sectors, centering on some sectors where the access time is not recovered thereafter, the distribution of the access delay for each sector is recorded, and it is shown in the figure. It is necessary to determine whether the number of sectors in which access delay is occurring is over the threshold value in the scatter plot zone of 7. As for this judgment, when focusing only on the occurrence of the access delay of the scatter diagram in FIG. 7, some problems obviously occur for the sector exceeding the alternative sector occurrence prediction threshold as in S501 of FIG. However, if an access delay occurs in an area exceeding the threshold shown in the scatter diagram of FIG. 7 below the alternative sector occurrence prediction threshold, a failure that actually leads to a failure with only the access delay time. It is difficult to determine if is occurring in that sector.

  Therefore, in addition to the fact that the access time of the command issue number i exceeds the threshold value, it is determined whether similar behavior can be seen in the sector adjacent to the command issue number i, and the progression of the failure reaching the life is predicted. This is because the access delay of each sector is not so great, but it is possible that a large number of access delays occur in the area near the command issue number i that the access completion of this area is not normal. It means that some kind of failure to extend the access time is in progress. In step S404, an area divided by the same number of sectors is used as the near segment area so that accurate determination can be performed using the number of sectors (command issue numbers) in which the access delay of the near area occurs. The predetermined number such as five segments before, five segments after, and ten or more warning counters in step S404 is an example, and a predetermined number other than the above-described values may be used.

  In step S406, when N command issue numbers in the segment to be accessed are viewed, if there is a predetermined number (for example, two) or more non-zero warning counters, a failure occurs that can not be recovered in the segment. In step S408, the error count is incremented by one, and the error counter value is passed to the error determination processing. At this point, since it is possible to identify the area in which the error has occurred, it is possible to extend the life of the HDD as a whole by taking measures such as avoiding the use of the problem area.

  As described above, in the warning determination process, it is possible to determine that a fault not counted as an error is accumulated in the HDD 300 by adding the warning counter. In addition, when a predetermined number or more of warning counters are generated in a certain segment area, error determination processing can be performed by immediately adding an error counter.

  FIG. 5 is a flowchart showing a detailed procedure of the error determination process of step S355. In the error determination process, it is determined how long the access time of the command issue number area in which the access time exceeds the threshold in the warning determination process of FIG. 4 actually requires the access time. Identify what anomalies you have.

  The error determination process is performed by adding the value of the error counter for each command issue number shown in FIG.

  In step S501, it is determined whether the access time exceeds the alternative sector occurrence prediction threshold. If the command execution time exceeds the total of the normal access times shown in FIG. 9A and the time for the abnormal processing determined in advance for each type of HDD 300 is significantly delayed, recovery processing is performed. It is likely to be in the alternative sector processing including. The processing time at the time of occurrence of an abnormality of the HDD 300 required for such alternative processing and the like is determined for each HDD 300. When it can be determined that such alternative sector processing is in progress (YES in S501), a state in which data is difficult to read is generated, so even if the warning counter is registered in the warning determination process of FIG. An error counter is added in S502.

  As described above, when the access time exceeds the alternative sector occurrence prediction threshold, if the number of occurrences is small and occurs alone, the access time at the time of subsequent re-access is recovered if the sector is shifted to the alternative sector. However, it is clear that there is a problem that obviously causes a problem in access at the location where the error occurred, so even if the warning counter is reset in step S407 in FIG. Do not initialize the counter by accumulating it as a counter.

  Next, in step S503, five or more non-zero error counters are confirmed in the second neighboring segment area in which ten segments before the segment to be accessed and ten segments later are taken (YES in S503) Then, it is determined that the access delay is expanded in the second neighboring segment area, and the process proceeds to step S505, where a warning of HDD stop is issued. This indicates that failure is concentrated in the second neighboring segment area. If errors occur continuously in the adjacent segment where LBA continues, a problem of write failure physically occurs in this area due to a scratch on the disk surface on the same circumference due to the rubbing of the head or a head failure. Because there is a high possibility.

  In such a case, it is highly likely that the physical fault has stayed in the command issue number area currently being evaluated, and has also spread to the command issue number area not yet evaluated, and a read error occurs during the evaluation process. Since there is a risk that data can not be read out, it is necessary to immediately issue an HDD stop warning and stop the use of the HDD.

  The number of segments in the adjacent second adjacent segment area of 10 before and 10 after the segment to be accessed and the number of 5 or more error counters in the second adjacent segment area are an example, and other than this The value of may be used. It is desirable to set an optimal value for each HDD 300 so that the smallest scratch can be found early on the innermost circumference of the disk with the fewest number of consecutive sectors in order to predict the wound of the largest disk one round early .

  In step S503, when the number of non-zero error counters in the second neighboring segment area is less than 5 (NO in S503), the process proceeds to step S504. Then, if the number of error counters of all the segments exceeds a predetermined value (for example, 10) in step S504 (YES in S504), an HDD stop warning is issued in step S505.

  As described above, according to the failure prediction procedure by the HDD failure prediction apparatus 100 of the present embodiment, failure prediction of the HDD 300 can be performed with high accuracy as described below, and data loss in the HDD 300 can be prevented.

  If an abnormal write due to particles being added below the head or a write error occurs accidentally when the head can not completely write data to the sector, the access time temporarily exceeds the threshold, but again By overwriting the same command issue number area, the access time is recovered, and thereafter it operates with the normal access time. In such a case, the warning counter is reset, and it can be determined that the HDD 300 has returned to normal operation.

  When the function is recovered due to the occurrence of the alternative sector, if it is single-ended, the error count does not advance since the function of the HDD 300 is normalized, but the alternative sector is continuously generated in the adjacent segment. It is considered that a writing error due to a scratch or a head failure has occurred. These are caused by progress of generation of an error counter in the segment before and after the problem sector or by addition of an error count considered as generation of a substitute sector if the access time exceeds a substitute sector generation expected threshold. , It can be determined that the failure of the HDD 300 is near.

  Further, the failure due to the life of the HDD 300 can be determined by checking whether the range in which the access time in which the warning counter is registered exceeds the threshold value is expanded before and after the specific segment. In this case, it can be judged visually from the fact that the scatter diagram of FIG. 7 is spreading compared to the normal time.

  In particular, if a predetermined number (for example, two) or more warning counters are generated in re-accessing the command issue number area in which the warning counter is registered, recovery due to a write error is not expected, and data write abnormality in this command issue number area is considered. Since the progress to the early failure can be considered, it is possible to more accurately judge that the occurrence of the failure is close by advancing the progress from the warning counter to the error counter.

  By accumulating the warning counter and shifting to the error counter in this manner, it is possible to more accurately determine the failure that progresses over time, in addition to the general failure determination of the HDD 300.

  Finally, the error counter is used to determine that the threshold for failure prediction has been exceeded from the accumulated value of the warning counter, and to notify the HDD 300 of a stop warning by an indicator light or the like. This may be an alarm by means of a buzzer or the like, and may be linked with the stop function of the present system. Also, in cooperation with an external system, display information related to the purchase of storage devices such as HDD (such as advertisement information) according to the value of warning counter or error counter, or use backup service using cloud etc. A courtesy service (eg, presentation of a coupon) may be provided to promote, promote the purchase of a storage device or the use of a backup service. As described above, by providing appropriate information to a highly necessary user pinpointedly, the convenience of the user is improved, and an increase in sales of related products and services can be expected.

  As described above, by deriving the threshold from the maximum access time during normal operation of the HDD 300, the range of the access time during normal operation of the HDD 300 can be known. Therefore, the abnormal operation is accurately performed in the command issue number area where the access time exceeds the threshold. The failure prediction of the HDD 300 can be performed with high accuracy.

  The present invention has been described above based on the embodiments. It is understood by those skilled in the art that the embodiment is an exemplification, and that various modifications can be made to the combination of each component and each processing process, and such a modification is also within the scope of the present invention. .

  In the above description, failure prediction is performed using the warning counter that may be reset and the error counter that is not reset, but the failure may depend on whether the warning counter exceeds a predetermined number using only the warning counter. May be determined.

  In the above description, the failure prediction technology has been described taking the hard disk as an example of the disk, but the failure prediction technology of the present embodiment can be applied to any magnetic disk or optical disk.

  10 HDD controller, 20 temporary storage unit, 30 control unit, 40 abnormal value DB recording unit, 100 HDD failure prediction device, 200 host, 300 HDD.

Claims (7)

An abnormal value recording unit that stores an upper limit threshold of an access time when an access target is accessed according to a predetermined access pattern for a disk drive;
When the measured value of the access time when the access target location is accessed according to the access pattern exceeds the upper threshold, status information related to the access target location is registered in the abnormal value recording unit, and the status information is stored in the status information. based on, viewed contains a control unit for abnormality detection,
The failure prediction apparatus, wherein the predetermined access pattern is a pattern for accessing an access target location after returning the head of the disk drive to an initial position . The failure prediction according to claim 1 , wherein the control unit detects an abnormality when the number of locations where the measured value of the access time exceeds the upper threshold exceeds a predetermined number in the vicinity area including the location to be accessed. apparatus. The control unit adds a warning counter corresponding to the access target location when the measured value of the access time exceeds the upper limit threshold at the access target location, and registers the result in the abnormal value storage unit. The warning counter is reset to zero and the abnormality is detected based on the warning counter when the access time is less than the upper threshold even when the access target location is registered. The failure prediction device according to claim 1 or 2 . The control unit is associated with the access target location when the number of the non-zero warning counters in the vicinity area including the access target location is equal to or greater than a predetermined number , and relates to abnormality detection more than the warning counter. The failure prediction apparatus according to claim 3 , wherein an error counter having a strong influence is added, and an abnormality is detected based on the error counter. According to the predetermined access pattern, the access times when accessing a plurality of access target locations are respectively measured, and using the access times of other access target locations located in the vicinity of the access target location, the access target location is The failure prediction apparatus according to any one of claims 1 to 4 , further comprising: a threshold value calculation unit that calculates the corresponding upper limit threshold value and causes the abnormal value recording unit to record the threshold value. Disk drive using device
And recording the abnormal value recording unit determines the upper limit threshold of the access time when accessing an access target portion by a predetermined access pattern for said disk drive,
When the measured value of the access time when the access target location is accessed according to the access pattern exceeds the upper threshold, status information related to the access target location is registered in the abnormal value recording unit, and the status information is stored in the status information. based on, look at including the, the method comprising the steps of abnormality detection,
The failure prediction method, wherein the predetermined access pattern is a pattern for accessing an access target location after returning the head of the disk drive to an initial position . Determining an upper limit threshold of access time when an access target location is accessed according to a predetermined access pattern for a disk drive and recording the upper limit threshold in an abnormal value recording unit;
When the measured value of the access time when the access target location is accessed according to the access pattern exceeds the upper threshold, status information related to the access target location is registered in the abnormal value recording unit, and the status information is stored in the status information. based on, a failure prediction program to be executed the steps of anomaly detection, to the computer,
The failure prediction program, wherein the predetermined access pattern is a pattern for accessing an access target location after returning the head of the disk drive to an initial position .

Images