JPH10283125A - Disk array device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve efficiency in cooling by enabling suitable cooling corresponding to the level of necessity in cooling for each disk device concerning a disk array device provided with a cooling device. SOLUTION: The temperature for each hard disk device 50 is estimated and based on the estimated temperature, it is discriminated for each hard disk device 50 whether cooling is required or not. Concerning the cooling device defining the hard disk device 50, for which unnecessity in cooling is judged, as a main cooling object, when there is any hard disk device, for which necessity in cooling is judged, among the hard disk devices 50 of auxiliary cooling objects positioned around that hard disk device 50 of main cooling object, the direction of wind is switched so as to apply the cooling operation to that hard disk device 50 but when there is no relevant device, driving is stopped. Therefore, suitable cooling is executed corresponding to the level of necessity in cooling for each hard disk device 50 and efficiency in cooling is improved.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a disk array device capable of writing and reading data to and from a predetermined disk device and used for a video server or the like which stores, for example, large-capacity video data.

[0002]

2. Description of the Related Art Conventionally, for example, RAID (Redundant
2. Description of the Related Art As represented by an array of ixpensive disks, a disk array device is known in which a plurality of disk devices are controlled by a disk array controller and data is written to and read from a predetermined disk device. A video server or the like that supplies a relatively large amount of data such as video data using the disk array device is configured.

In this disk array device, in order to prevent the performance of reading data from deteriorating, many of the plurality of disk devices provided are operated so that reading can be started at any time. Have been. During the operation of the disk device, a large amount of heat is generated due to the operation of the disk drive spindle motor and the like, and the generated heat may cause a failure. In particular, in a disk array device, many disk devices are often arranged very close to each other, and when a high-speed rotating disk device is used to improve performance, heat generation is particularly severe.
For this reason, a fan device for cooling the disk device is provided, for example, for each unit, and while the power is turned on to the disk array device, the fan device is operated to suppress a rise in temperature.

[0004]

However, even if all the disk devices are operated as described above, it is not always necessary to read data from all the disk devices, but to actually read data. Often a part of them. Of course, even though the disk device to be read varies depending on the type of data, the disk device to be read at the same time is often a part. Therefore, for example, even if the video server system is operated for 12 hours, there is a possibility that a disk device in which data is read out for only about one hour or a disk device in which data is not read out even once during operation may appear. There is also.

In other words, among a large number of disk devices, there are variations in the operating status, such as those that are not operating, those that are operating but not being accessed, and those that are operating and being accessed. There is. Further, there are cases where the number of errors suddenly increases due to a rise in temperature, and cases where the number of errors does not increase. In other words, there are those that need cooling at each time and those that do not particularly need to be cooled, and if all of them are cooled in the same manner, the cooling efficiency deteriorates.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and in a disk array device having a cooling device, appropriate cooling can be performed according to the degree of cooling required for each disk device. The purpose is to improve the cooling efficiency.

[0007]

According to the first aspect of the present invention, there is provided a disk array controller which controls a plurality of disk units by using a disk array controller and transmits data to a predetermined disk unit. In a disk array device capable of writing and reading, a cooling device capable of providing a cooling function to the disk device is arranged such that a cooling device corresponding to each disk device is provided, and the cooling device is The disk array controller is configured to be switchable to a state in which a cooling action can be applied to the auxiliary cooling target disk device disposed near the corresponding main cooling target disk device, and the disk array controller controls the temperature of each disk device. Device temperature estimating means for estimating the temperature and the data based on the temperature estimated by the device temperature estimating means. The cooling necessity determining means for determining whether or not cooling is necessary for each disk device, and the cooling device whose main cooling target is a disk device determined to be unnecessary by the cooling necessity determining means, To stop or
Alternatively, if there is a disk device to be subjected to auxiliary cooling that is determined to require cooling, a cooling state control means for switching to a state in which a cooling effect is applied to the disk device is provided.

According to the disk array device of the present invention, the cooling devices capable of providing a cooling function to the disk devices are arranged such that a corresponding cooling device exists for each disk device. The meaning of “there is a corresponding cooling device for each disk device” naturally applies, for example, if one cooling device is arranged for each disk device. With respect to two disk devices, a configuration in which one cooling device is in charge of cooling is also conceivable.

The cooling device is configured to be switchable to a state in which a cooling effect can be given to a disk device to be auxiliary cooled, which is arranged near a corresponding disk device to be main cooled. For example, consider a disk array configuration in which a plurality of rows of disk devices are arranged in a plurality of rows in the left-right direction, and in order to simplify the description, one cooling unit is provided in front of one disk device. Consider the case where the device is arranged as an example. In this case, for the cooling device, the disk device located at the front corresponds to the “main cooling target disk device”, and the disk device arranged adjacent to the disk device in the up, down, left, and right directions, respectively. This corresponds to “a disk device to be subjected to auxiliary cooling”. As a matter of course, for the cooling device that is in charge of the disk device located at the end of the device array as the main cooling target, the right side exists as the auxiliary cooling target disk device, but the left side does not exist. Conceivable.

On the premise of such a configuration, the disk array controller estimates the temperature of each disk device,
The necessity of cooling is determined for each disk device based on the estimated temperature. For a cooling device that is set as a main cooling target for a disk device that is determined to be unnecessary for cooling, the drive of the cooling device is stopped, or a cooling device that is determined to need cooling for a disk device to be auxiliary cooled is determined. If there is, the state is switched to a state in which the disk device is provided with a cooling action.

As described above, the disc lay device of the present invention is provided with the cooling device for giving a cooling effect to the disc device, and performs appropriate cooling according to the degree of cooling required for each disc device. And cooling efficiency can be improved. In addition, as described in claim 2, the cooling state control means, when it is determined that cooling is required in the auxiliary cooling target disk device,
Only when the number of cooling devices that provide the cooling function to the disk device is equal to or less than a predetermined number, the state may be switched to the state where the cooling function is provided. For example, if it is sufficient that the cooling action from two cooling apparatuses is provided, even if a third corresponding cooling apparatus is present, the state is switched to the state where the cooling action is provided. Without stopping.

Although there are various methods for generating a cooling action of the cooling device, the cooling action is given to the disk device by blowing air, and the auxiliary cooling is performed by adjusting the blowing direction. It is common to switch to a state in which a cooling effect can be applied to the target disk device. Of course, a water cooling method can be realized instead of the air cooling method.

According to the present invention, the temperature of each disk device is estimated, and the necessity of cooling is determined for each disk device based on the estimated device temperature.
As described in claim 4, it is conceivable to estimate the device temperature based on the operation status of the disk device. Further, when the apparatus temperature is estimated based on the operating state, as described in claim 5, the operating time during which the disk apparatus is in a state where data can be written and read, and the data It is conceivable to estimate the device temperature based on the access time during which the writing and reading of the data are performed.

The operating time in this case indicates that the disk device is in a state in which data can be written and read, and more specifically, a state in which the disk is rotated by a spindle motor. I do.
Therefore, this rotation driving operation itself causes heat generation. In addition, during the access time in which the writing and reading of data are performed, a predetermined data process of writing and reading data by driving the reading / writing head while the disk is rotating is executed. In this case, a head driving operation or the like is further added, which further causes heat generation. Therefore,
Appropriate estimation is possible by estimating the device temperature based on the time during which such a heat generation state continues.

It is conceivable that the degree of heat generation differs depending on the type of disk device even if the operating time and access time are the same. Therefore, it is conceivable that the device temperature is estimated based on the operating time and the access time in consideration of the influence of the difference in the type of the disk device on the device temperature estimation.

In the estimation, the correspondence between the operating time, the access time, and the temperature rise value is stored for each disk device model.
The temperature rise value corresponding to the operating time and the access time of the disk device to be estimated is read out from the stored relationship,
It is conceivable to estimate the device temperature based on the temperature rise value. This correspondence may be obtained in advance by an experiment or the like.

The factors that affect the device temperature include the temperature of the environment in which the disk devices are arranged, in addition to the above-described operation status of the device itself. Therefore, as set forth in claim 8, an environmental temperature estimating means for estimating the environmental temperature may be further provided, and the apparatus temperature may be estimated based on the estimated environmental temperature. Regarding the estimation of the environmental temperature, for example, the temperature of the installation location of the disk device may be actually measured to estimate the temperature of the installation location, or the correspondence between the time and the environmental temperature may be sampled and stored in advance. When estimating the disk device temperature,
The environmental temperature corresponding to the time may be estimated from the correspondence.

In the above-mentioned disk array apparatus, the present invention further comprises an error rate detecting means for detecting an error rate for each disk apparatus, and the cooling necessity determining means is provided by the error rate detecting means. The necessity of cooling for each disk device may be determined based on the detected error rate.

In this determination of the necessity of cooling, for example, it is considered that cooling is necessary if the error rate is larger than a predetermined threshold value, and that cooling is unnecessary if the error rate is less than the predetermined threshold value. It is preferable to give priority to the determination based on the device temperature. That is, if the device temperature is higher than a predetermined high-temperature side threshold, it is determined that the operation of the cooling device is unconditionally necessary, and if the device temperature is equal to or lower than the predetermined low-temperature side threshold, the operation of the cooling device is unconditionally unnecessary. Should be determined as
If it is between the two threshold values on the high temperature side and the low temperature side, the determination is first made based on the error rate.

One of the reasons based on the error rate is that even if the temperature of the apparatus becomes high similarly,
In some devices, the error does not increase so much and the adverse effect of high temperature is small, and in other devices, the error increases rapidly and the adverse effect of high temperature is considerable. Therefore, the necessity of cooling can be more appropriately determined based on not only the device temperature but also its error rate. However, an increase in the error rate does not always correspond to an increase in the temperature of the apparatus.For example, if the cumulative operating time becomes extremely long, an error may occur due to deterioration in a driving part. There is also. Therefore, based on the degree of increase in the error rate corresponding to the change in the apparatus temperature, it may be determined that an apparent increase in the apparatus temperature causes an increase in errors.

[0021]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of the disk array device 1. As shown in FIG. 1, the disk array device 1 of the present embodiment includes a disk array controller 10 that controls the entire disk array device 1, a plurality of hard disk devices 50 that are controlled by the disk array controller 10, And a cooling device 70 to be used.

The disk array controller 10 serves as control means corresponding to "apparatus temperature estimating means", "cooling necessity determining means", "cooling state control means", "environmental temperature estimating means", and "error rate detecting means". The CPU 30 is mainly configured.
ROM 31 storing a program to be executed by
A RAM 32 which is a work area of the PU 30 and corresponds to a "correspondence storage means", an interface (I / F) 35 as a communication means for transmitting and receiving data to and from a host device (not shown), and a hard disk device 50 will be described later. Drive control device 40 for controlling in system units
And are connected.

A plurality of hard disk devices 50 are connected to each drive control device 40 in parallel by a bus line 55. In this embodiment, one drive row is composed of a hard disk devices 50, and this drive row is b rows. In FIG. 1, for identification of each hard disk device 50, the system number is represented by a and the drive column number is represented by b, which is represented by (a, b).

The system number and the drive row number are each 0
, The number of systems is (a +
Since (b + 1) drive rows exist in 1), a total of (a + 1) × (b + 1) hard disk devices 50 are housed in one housing (not shown). In this case, when the cooling device 70 is not provided, the temperature of the hard disk device 50 rises considerably during long-time continuous operation, and especially when the high-speed rotating hard disk device 50 is used to improve performance, the temperature rises. Is intense. Therefore, in the present embodiment, one cooling device 70 is provided for each hard disk device 50.

FIG. 2 is a perspective view showing a schematic configuration of the cooling device 70. The cooling device 70 has a blower fan 71
Are arranged, and there are fins for adjusting the wind direction in two stages in front of the blowing direction. The fins 73 for adjusting the vertical direction are located farther from the blower fan 71, and the fins 75 for adjusting the horizontal direction are located closer to the fan. These adjusting fins 73 and 75 are provided in the disk array controller 1.
0 so that the wind direction can be freely changed in each of the up, down, left, and right directions. The cooling device 70 is arranged so as to face the individual hard disk devices 50, and as described above,
In the case of the present embodiment, one hard disk drive 50 adjacent to each of the three hard disk drives 50 can be cooled by changing the wind direction of each of the hard disk drives 50 in the vertical and horizontal directions. The cooling device 7
For the hard disk drive 50, the corresponding hard disk drive 50 corresponds to the "main cooling target disk drive", and the hard disk drive 50 disposed adjacent to the hard disk drive 50 in the vertical and horizontal directions, respectively. Corresponds to the “disk device to be subjected to auxiliary cooling”.

Next, the operation related to the control of the cooling device 70 in the disk array device 1 of the embodiment configured as described above will be described. FIG. 3 is a flowchart showing a procedure of the cooling device operation necessity determination process executed by the CPU 30 of the disk array controller 10.

In the first step S10, a hard disk temperature calculation process is executed. This processing will be described later with reference to FIG. In the following S20, the hard disk temperature hdd_temp [route] [d
rv] is higher than the high temperature side threshold value TEMP_HIGH_CONST.

Then, the hard disk temperature hdd_te
mp [route] [drv] is the high temperature side threshold TEMP_
If it is larger than HIGH_CONST (S20: YE
S), cooling device operation necessity flag fan_move [ro
ute] [drv] is set to "1" indicating that cooling is necessary (S30), and the cooling device operation necessity determination processing routine ends.

On the other hand, the hard disk temperature hdd_tem
p [route] [drv] is the lower threshold TEMP_L
If it is larger than OW_CONST (S20: NO, S3
0: YES), error rate err_ratio [rout
e] [drv] is the error threshold ERR_LEVEL_CO
It is determined whether it is larger than NST. And the error rate err
_Ratio [route] [drv] is the error threshold E
If it is larger than RR_LEVEL_CONST (S6
0: YES), cooling device operation necessity flag fan_mov
e [route] [drv] is set to “1” indicating that cooling is necessary (S30), and the error threshold ERR is set.
_LEVEL_CONST or less (S60: N
O), cooling device operation necessity flag fan_move [ro
ute] [drv] is set to "0" indicating that cooling is unnecessary (S70), and the cooling device operation necessity determination routine ends.

The hard disk temperature hdd_tem
p [route] [drv] is the lower threshold TEMP_L
If OW_CONST or less (S20: NO, S5
0: NO), cooling device operation necessity flag fan_move
[Route] [drv] indicates “0” indicating that cooling is unnecessary
Is set (S70), and the cooling device operation necessity determination routine is terminated.

According to the above processing, if the hard disk temperature hdd_temp [route] [drv] is higher than the high temperature side threshold TEMP_HIGH_CONST,
It is determined that the operation of the cooling device 70 is unconditionally necessary, and if it is equal to or lower than the low-temperature-side threshold value TEMP_LOW_CONST,
It is unconditionally determined that the operation of the cooling device 70 is unnecessary.
If the error rate err_ratio [route] [drv] is larger than the error threshold ERR_LEVEL_CONST between the two thresholds on the high temperature side and the low temperature side, it is determined that the operation of the cooling device 70 is necessary,
When the error is equal to or less than the error threshold ERR_LEVEL_CONST, it is determined that the operation of the cooling device 70 is not.

Next, details of the hard disk temperature calculation processing executed in S10 of FIG. 3 will be described with reference to the flowchart of FIG. In the first step S100 of this processing routine, the hard disk drive 50
Issue a predetermined SCSI command to acquire the model of the hard disk device 50. The head address of the table corresponding to the hard disk model of the temperature rise value table is set to the table pointer tbl_pointe.
r (S110).

This temperature rise value table is a table in which the temperature rise values for each model of the hard disk device 50 obtained in advance by experiment are associated with the operation time and the operation rate, and the data is changed by an external input. Is also possible. The operating time refers to the time during which power is supplied and access such as data reading and writing is possible, and the operating rate is the ratio of the actual access time to the operating time. It is.

In the following S120, the current time is obtained, and S
At 130, environmental temperature data corresponding to the current time is read from the environmental temperature table, and is read out as environmental temperature env_t.
Store in emp. Here, the environmental temperature table is a table in which one day is divided into predetermined time units, and the time and the environmental temperature are associated on a one-to-one basis, and data can be changed by an external input.

In S140, an offset value of the temperature rise value table is obtained from the operation time t_sum and the operation rate t_move_ratio, and this is calculated as the offset value offset.
To memorize. Hard disk temperature hdd_temp [r
out] [drv] is the table pointer tbl_p
Since the result is obtained by adding a value obtained by multiplying the environmental temperature by a predetermined coefficient to the temperature rise value read from the position shifted by the offset value offset from the pointer, in S150, this is stored in the hard disk temperature temp, and thereafter, Then, the hard disk temperature calculation processing routine ends.

By the above-described processing, the hard disk temperature is calculated based on the four factors of the hard disk model, the environmental temperature, the operation time, and the operation rate. Next, the contents of the hard disk access processing will be described with reference to the flowchart of FIG.

In the first step S200, the current time is obtained and stored in t_now. This time t_
Since now is the start time of the current access to the hard disk device 50, in S210, this is stored in the access start time t_start.

The time required from the current time t_now to the time t_end at which the previous access ended is
Since there is no access after the previous access, the access end time t_en is changed from the current time t_now.
The value obtained by subtracting d is stored in the access non-execution time t_idle (S220).

Subsequently, this value is added to the total access non-execution time to update the access non-execution time (S23).
0). Then, the access number access_count
Is incremented (S240), and access is executed (S250). When the access is completed, it is determined whether the access was an error (S260).
260: YES), the number of errors err_count is incremented (S270).

Since the value obtained by dividing the number of errors err_count by the number of accesses access_count is the error rate, this is calculated as the error rate err_ratio.
(S280). Next, the current time is acquired and stored in t_now (S290). Since this time is the end time of the current access, this time is referred to as t_
The end is stored (S300).

Since the value obtained by subtracting the access start time t_start from the current time t_now is the time during which the access was performed this time, this value is stored in t_move (S310). Total access execution time t_move_
This value is added to sum, and the access execution time t_move
Is updated (S320).

Since the operating time is eventually the sum of the access execution time and the access non-execution time, the total access execution time t_move_sum is added to the total access non-execution time t_idle_sum, and the result is stored in the total operation time t_sum. (S330).

Since the operation rate is the ratio of the access time to the operation time, the total access time t_mov
e_sum is divided by the total operation time t_sum, and the result is stored in the operation rate t_move_ratio (S34).
0), the hard disk access processing routine ends.

By this processing, the hard disk drive 5
Each time 0 is accessed, the error rate err_ratio is updated. Next, the content of the cooling device control processing will be described with reference to the flowchart of FIG.

In the first step S400, the drive row number drv is initialized to 0. If the drive row number drv is larger than the drive row number maximum value MAX_DRV (S410: YES), all the hard disk drives 5
Since the check of the operation necessity check result of 0 has been completed, the fan operation determination processing is executed (S480), and the cooling device control processing routine ends (S490). Note that S48
The fan operation determination processing routine executed at 0 will be described later with reference to FIG.

On the other hand, if the drive row number drv is equal to or less than the drive row number maximum value MAX_DRV (S410: N
O) Since the check has not been completed, the processing after S420 is executed to continue the check. In S420, the system number route is initialized to 0. And the system number r
out is the maximum value of the system number MAX_ROUTE (= 4)
If it is larger (S430: YES), the drive row check is completed, and the drive row number drv is incremented (S500). After that, the process shifts to S410 to continue checking.

On the other hand, if the system number route is equal to or less than the system number maximum value MAX_ROUTE (S430: N
O) Since the check of this drive row has not been completed, the check is continued. If fan operation is required (S
440: YES), 1 is stored in fan_count [route] [drv] for storing the number to which the fan is directed (S450), and fan_a for storing the direction of the fan.
S that means the front in ngle [route] [drv]
TRAIGHT is set (S460), and the system number ro
ute is incremented (S470). Then, S
Proceed to 430 to continue checking.

If there is no fan operation (S440: N
O), fan_co that stores the number to which the fan is directed
0 is stored in unt [route] [drv] (S51)
0), fan_angle [r that stores the direction of the fan
out] [drv] is set to OFF meaning stop (S520), and the system number route is incremented (S470). Thereafter, the process proceeds to S430 and the check is continued.

For those requiring fan operation, STRAIGHT is set so that the fan angle is directed to itself, and the fan itself is cooled. In this case, STOP is set because there is a possibility of stopping the fan for the time being, and the number to which the fan is directed is set to 0 because it does not cool itself.

Next, details of the fan operation determination processing executed in S480 of FIG. 6 will be described with reference to the flowchart of FIG. In the first step S600 of this processing routine, the drive row number drv is initialized to 0 (S600), and if the drive row number drv is larger than the drive row number maximum value MAX_DRV (S61).
0: YES), since the fan operations of all the hard disk devices 50 have been determined, the present fan operation determination processing routine ends.

On the other hand, if the drive row number drv is equal to or less than the drive row number maximum value MAX_DRV (S610: N
O) Since there is a hard disk for which the fan operation has not been determined, the processing after S620 is executed to continue the fan operation determination processing. In S620, the system number route is initialized to 0 (S620). If the system number route is larger than the system number maximum value MAX_ROUTE (= 4) (S630: YES), all the fan operations of the drive row are determined. The column number drv is incremented (S720). After that, the flow shifts to S610, where the fan operation determination processing is continued.

If the system number route is equal to or less than the system number maximum value MAX_ROUTE (S630: N
O) Since there is a hard disk drive 50 for which the fan operation has not been determined in the drive row, the flow shifts to S640 to continue the present fan operation determination process.

If fan operation is required (S640: N
O), since the hard disk device 50 has already determined the fan direction to cool itself, the process proceeds to S670. On the other hand, if the fan operation is unnecessary (S640: Y
ES), the fan direction is off for the time being, but if there is a hard disk device 50 that requires a fan operation on its own top, bottom, left and right, the fan is directed to this hard disk device 50 to operate the fan.
In steps 0, S660, S730 to S780, the states of the upper, lower, left, and right hard disk devices 50 are checked.

In S650, upper side check processing is performed.
When the direction is determined by the upper side check processing (S660:
If YES, the process proceeds to S670, but if not determined (S66)
0: NO), a lower side check process is performed (S730). S
If the direction is determined by the lower side check process at 730 (S740: YES), the process proceeds to S670, but if not determined (S740: NO), the left side check process is performed (S740).
760).

If the direction is determined by the left-side check process in S760 (S770: YES), the process proceeds to S670, but if not (S770: NO), the right-side check process is performed (S780) and the process proceeds to S670. Note that S6
The "upper check process" at 50 is the flowchart of FIG. 8, the "lower check process" at S730 is the flowchart of FIG. 9, the "left check process" at S760 is the flowchart of FIG.
The "right side check process" at 80 will be described later in detail with reference to the flowchart of FIG.

In S670 of FIG. 7, the fan direction fan_angle [route] determined in the processing up to that point is performed.
e] Change the direction of the fan according to [drv]. Then, the fan direction fan_angle [route] [d
If [rv] is OFF (S680: YES), the fan is stopped (S740). On the other hand, the fan direction is OFF
If not (S680: NO), the fan is operated (S690).

After the processing in S690 or S750, the system number route is incremented (S700), and S6
The process proceeds to 30 to continue the fan operation determination processing routine. Through the above-described processing, the direction in which the fan is directed and whether to operate or stop the fan are determined.

Next, details of the upper-side check process executed in S650 of FIG. 7 will be described with reference to the flowchart of FIG. In the first step S800 of this processing routine, it is determined whether or not the system number route is 0, and if the system number route is 0 (S
800: YES), the upper hard disk device 50 does not exist, so UN-DECIDED meaning undecided is set to the return value ret (S870), and the upper check process routine ends.

On the other hand, if the system number route is not 0 (S800: NO), it means that the upper hard disk device 50 exists, and a predetermined check process is performed.
First, it is determined whether the upper hard disk device 50 needs a fan operation (S810). In the case of the present embodiment, the upper hard disk drive 50 has the same drive row number as the previous system number, so the determination in S810 is based on the cooling device operation necessity flag fan
The determination is made based on whether _move [route-1] [drv] is 1. If a negative determination is made here (S8
10: NO), since it is not necessary to direct the fan to the hard disk device 50, UN-DECIDED meaning undecided is set to the return value ret (S870), and the upper-side check processing routine ends (S860).

On the other hand, if the upper hard disk device 50 needs a fan operation (S810: YES), the process proceeds to S8.
Then, it is determined whether or not two fans are already facing the hard disk device 50. This determination is based on the number of fans fan_count [route] [d
rv] is 2. And S82
0, that is, if two fans are already facing, a new fan is added to the hard disk drive 50.
Therefore, UN-DECIDED meaning undecided is set as the return value ret (S870), and the upper-side check processing routine ends.

On the other hand, if a negative determination is made in S820, that is, if two fans are not facing, the process proceeds to S830, and
Fan angle fan_angle [route] [dr
v] is set to UP. Then, the return value ret is set to DECIDED meaning determination (S
840), the number of the upper hard disk drive 50 to which the fan is directed is incremented (S85).
0), the upper check routine ends.

By this processing, the hard disk drive 50 above the hard disk drive 50 that does not require a fan operation.
If the fan operation is required, and if only one of the fans originally installed in the hard disk device 50 is suitable for the hard disk device 50, the fan that does not need the fan operation is replaced with the hard disk device that requires the fan operation. Device 5
Decide to turn to 0.

FIG. 9, FIG. 10, and FIG. 11 are flowcharts showing the procedures of the lower-side check processing, the left-side check processing, and the right-side check processing, respectively. These processes are basically similar to the upper-side check process described with reference to the flowchart in FIG. 8, and thus detailed description is omitted. That is, only the predetermined processing is different in relation to the position of the target hard disk device 50.

For example, in the case of the upper-side check process in FIG. 8, in the first step S800, the system number route
Is determined to be 0, whereas in the case of the lower-side check process in FIG. 9, in the first S900, it is determined whether or not the system number route is the system number maximum value MAX_ROUTE. Then, in the case of the upper side check processing of FIG.
Therefore, the system number is [route-1] one before and the drive row number is the same [drv]. However, in the case of the lower check process in FIG. Therefore, the system number is shifted by one [ro
ute + 1], the drive row number becomes the same [drv]. Needless to say, in the case of FIG. 8, the fan angle fan_angle [route] [drv] is set to UP in S <b> 830, but in FIG. 9, of course, DOWN is set to mean lower.

Further, the left-side check processing in FIG.
In the right-side check process of 1, in the first judgment, it is determined whether the drive row number drv is 0 (S1000 in FIG. 9) or the drive row number maximum value MAX_DRV (S1100 in FIG. 10). Since the left and right hard disk devices 50 have the same system number and the drive row number is one before or one after, the drive row number is set to [drv-1] or [drv + 1], and various determination / setting processes are performed. . Of course, in this case, the fan angle fa
n_angle [route] [drv] is set to LEFT meaning left as shown in S1030 of FIG. 10 or RIGHT meaning right as shown in S1130 of FIG.

Although the overall control process is not specifically shown, the cooling device operation necessity determination process shown in FIG. 3 is performed in predetermined time units, and thereafter, the cooling device control process shown in FIG. Hard disk temperature of the hard disk device 50 (hdd_temp [route] [d
rv]) and the error rate (err_ratio), the cooling device 70 for the hard disk that does not require cooling is operated toward the hard disk device 50 that requires cooling, or the driving of the cooling device 70 itself is stopped if unnecessary. Or let it. Thus, the present disk array device 1
In the above, the temperature of each hard disk device 50 is estimated, and the hard disk device 5 is estimated based on the estimated temperature.
The necessity of cooling is determined for each 0. Then, for the cooling device 70 whose main cooling target is the hard disk device 50 that is determined to be unnecessary for cooling, the drive of the cooling device 70 is stopped or it is determined that cooling is necessary for the hard disk device 50 that is the auxiliary cooling target. If any of them has been performed, the hard disk device 50 is switched to a state in which a cooling action is provided. Therefore, appropriate cooling can be performed according to the degree of cooling required for each hard disk device 50, and cooling efficiency can be improved.

The estimation of the device temperature is also based on the environmental temperature and the operating status of the hard disk device 50. This operating state is distinguished into an operating time during which data can be written and read, and an access time during which data writing and reading have been performed. That is, based only on the operating time, in the hard disk drive 50, the magnetic disk is rotated by the spindle motor but the magnetic head is not driven in some cases, and the magnetic head is also driven. There is also a state that is. Therefore, in consideration of the fact that the rotational drive operation of the magnetic disk and the drive operation of the magnetic head each cause heat generation, it is possible to perform more appropriate estimation by distinguishing these operations.

Further, in the disk array device 1 of the present embodiment, the necessity of cooling is determined based on the error rate of the hard disk device 50. In this cooling necessity determination, it is determined that cooling is necessary if the error rate is greater than a predetermined threshold, and that cooling is unnecessary if the error rate is less than the predetermined threshold, but priority is given to determination based on the estimated device temperature. Let me. That is, if the device temperature is higher than a predetermined high-temperature side threshold, it is determined that the operation of the cooling device is unconditionally necessary, and if the device temperature is equal to or lower than the predetermined low-temperature side threshold, the operation of the cooling device is unconditionally unnecessary. Is determined between the two threshold values on the high temperature side and the low temperature side.

As described above, one of the reasons based on the error rate is that even if the device temperature becomes high similarly, the error does not increase so much due to the individual difference of the hard disk device 50 and the adverse effect due to the high temperature. For some devices, the error increases sharply and the adverse effects of high temperatures are substantial. Therefore, based not only on the device temperature but also on its error rate,
The necessity of cooling can be more appropriately determined. However, an increase in the error rate does not always correspond to an increase in the temperature of the apparatus.For example, if the cumulative operating time becomes extremely long, an error may occur due to deterioration in a driving part. There is also. Therefore, based on the degree of increase in the error rate corresponding to the change in the apparatus temperature, it may be determined that an apparent increase in the apparatus temperature causes an increase in errors.

As described above, the present invention is not limited to such an embodiment, and can be implemented in various forms without departing from the gist of the present invention. Some of them are listed below. In the above-described embodiment, the configuration is such that the preset data of the hard disk temperature rise value and the environmental temperature are read out as table data. However, the configuration may be such that these data are obtained directly from a calculation formula derived based on the experimental results. Conceivable.

It is not necessary to have the temperature rise value tables as many as the types of the hard disk devices 50. Depending on the degree of temperature rise, many types may be divided into several groups to reduce the number of tables. It is not necessary to acquire the model of the hard disk device 50 every time, but at the time of starting up the system. There is no problem even if the result is stored and referred to.

There is no problem even if the error rate is not observed at all and the determination is made based only on the estimated device temperature. It has been described that the hard disk device 50 that needs cooling is provided with a cooling effect from a maximum of two cooling devices 70, but three or more cooling devices may be provided.

The direction of changing the wind direction by adjusting the adjusting fins 73 and 75 is such that the hard disk devices 50 adjacent to the upper, lower, left and right sides of the hard disk device 50 to be main cooled are directed as auxiliary cooling objects. The hard disk devices 50 located at other positions such as “on the upper left side of the hard disk device 50” and one hard disk device 50 above may be subjected to auxiliary cooling, and the direction of them may be changed.

The number of hard disk temperature thresholds may be increased to change the upper limit of the number of hard disk drives 50 to which the thresholds are directed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a disk array device according to an embodiment.

FIG. 2 is a schematic perspective view illustrating a configuration of a cooling device according to the embodiment.

FIG. 3 is a flowchart illustrating a cooling device operation necessity determination process executed by a CPU of the disk array controller.

FIG. 4 is a flowchart showing a hard disk temperature calculation process executed by the CPU of the disk array controller.

FIG. 5 is a flowchart showing a hard disk access process executed by the CPU of the disk array controller.

FIG. 6 is a flowchart showing a cooling device control process executed by the CPU of the disk array controller.

FIG. 7 is a flowchart showing a fan operation determination process executed by the CPU of the disk array controller.

FIG. 8 is a flowchart showing an upper-side check processing routine executed during the fan operation determination processing.

FIG. 9 is a flowchart illustrating a lower-side check processing routine executed during the fan operation determination processing.

FIG. 10 is a flowchart illustrating a left-side check processing routine executed during the fan operation determination processing.

FIG. 11 is a flowchart illustrating a right-side check processing routine executed during the fan operation determination processing.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Disk array device 10 ... Disk array controller 30 ... CPU 31 ... ROM 32 ... RAM 40 ... Drive control device 50 ... Hard disk device 55 ... Bus line 70 ... Cooling device 71 ... Ventilation fan 73 ... Vertical adjustment fins 75 ... Right and left Direction adjustment fins

Claims (9)

[Claims] In a disk array device in which a plurality of disk devices are controlled by a disk array controller and data can be written to and read from a predetermined disk device, a cooling device capable of providing a cooling function to the disk device is provided. ,
The cooling devices are arranged such that a corresponding cooling device is provided for each of the disk devices, and the cooling device imparts a cooling effect to the auxiliary cooling target disk devices arranged near the corresponding main cooling target disk device. The disk array controller is configured to be switchable to a possible state, wherein the disk array controller includes: a device temperature estimating unit for estimating a temperature of each of the disk devices; and a temperature estimated by the device temperature estimating unit.
A cooling necessity determining means for determining necessity of cooling for each of the disk devices; and a cooling device whose main cooling target is a disk device determined to be unnecessary by the cooling necessity determining means. Cooling state control means for stopping the drive or switching to a state in which a cooling action is applied to the disk device if any of the disk devices to be cooled is determined to require cooling. A disk array device characterized by the above-mentioned. 2. The cooling device according to claim 1, wherein said cooling state control means applies a cooling function to said disk device when a disk device to be cooled is determined to require cooling. 2. The disk array device according to claim 1, wherein switching is performed to a state in which a cooling effect is applied only when the number is equal to or less than a predetermined number. 3. The cooling device can apply a cooling effect to the disk device by blowing air, and adjusts a blowing direction to a state in which a cooling effect can be applied to the disk device as the auxiliary cooling target. 3. The disk array device according to claim 1, wherein the disk array device is configured to be switchable. 4. The disk array device according to claim 1, wherein said device temperature estimating means estimates a device temperature based on an operation state of said disk device. 5. The device temperature estimating means includes: an operating time during which the disk device is in a state where data can be written and read; and an access time during which data is written and read from the disk device. 5. The disk array device according to claim 4, wherein the device temperature is estimated based on the following. 6. The apparatus temperature estimating means estimates the apparatus temperature based on the operating time and the access time, taking into account the effect of the difference in the type of the disk device on the apparatus temperature estimation. The disk array device according to claim 5, wherein 7. A correspondence storage means for storing the correspondence between the operating time and the access time and the temperature rise value for each model of the disk device, wherein the device temperature estimating means comprises: 7. The disk array device according to claim 6, wherein a temperature rise value corresponding to the operation time and access time of the disk device is read out from the correspondence storage means, and the device temperature is estimated based on the temperature rise value. . 8. An environment temperature estimating means for estimating a temperature of an environment in which the disk device is disposed, wherein the apparatus temperature estimating means is configured to perform an operation based on the environmental temperature estimated by the environmental temperature estimating means. 8. The disk array device according to claim 1, wherein the temperature is estimated. 9. An error rate detecting means for detecting an error rate of each of the disk devices, wherein the cooling necessity determining means determines the error rate of each of the disk devices based on the error rate detected by the error rate detecting means. 9. The necessity of cooling is determined.
The disk array device according to any one of the above.