JPH1196083A - Semiconductor disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a semiconductor disk device capable of preventing the destruction/disappearance of data stored in a semiconductor memory by normally executing the restoring processing of data at the time of turning on a power supply even when a defect exists on a backup hard disk drive(HDD) medium. SOLUTION: A control part 11 saves data stored in a memory 12 to plural different saving areas previously formed in a backup HDD 13 in accordance with the interruption of power supply from a power source 19. When power supply from the power source 19 is started, the control part 11 restores the data stored in a prescribed saving area to the memory 12. When the restoring processing is not normally executed, the control part 11 restores data stored in a saving area different from the former saving area to the memory 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a semiconductor disk device using a volatile memory as a storage medium, and more particularly to a semiconductor device having a hard disk (HDD) for backing up data stored in the volatile memory. It relates to a disk device.

[0002]

2. Description of the Related Art In recent years, in a computer system such as a personal computer, a semiconductor disk device capable of using a semiconductor memory as a storage medium and reading / writing data using tracks and sectors as well as accessing a hard disk (HDD). Is attracting attention. Such a semiconductor disk device has been widely used in a conventional H disk.
There is an advantage that data can be read / written at a higher speed than DD.

However, when a volatile memory such as a DRAM (dynamic random access memory) is applied to the storage medium of such a semiconductor disk device, the data is stored in the semiconductor memory by turning off the power of the semiconductor disk device. There is a problem that the data that has been lost will be lost. For this reason,
2. Description of the Related Art A semiconductor disk device having a backup HDD for holding data stored in a semiconductor memory while power is cut off is provided.

In such a semiconductor disk device, in response to a power-off command, power-off is delayed for a predetermined time, or data stored in a semiconductor memory is backed up by power from a battery or the like provided in advance. A process of saving (saving) to the HDD is performed. When the power is turned on, a restore process for returning the saved data to the semiconductor memory is performed.

[0005]

However, in the above-mentioned semiconductor disk device, when the backup HDD has a defect, there occurs a problem that the data saved at the time of turning on the power is not normally restored in the semiconductor memory.

In the semiconductor disk device, FIG.
As shown in (2), for example, even if there is a defect on the medium of the backup HDD, the data saving process according to the power-off command is performed normally. As described above, when the data saving process is performed as usual even though the medium has a defect, the next time the power is turned on, as shown in FIG. The restore process cannot be performed completely, and the semiconductor disk device cannot be started up normally.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned circumstances. Even when a defect is found on a medium of a backup HDD, a data restoration process at the time of power-on is normally executed, and a semiconductor memory is provided. It is an object of the present invention to provide a semiconductor disk device capable of preventing the destruction / loss of data stored in a storage device.

[0008]

A semiconductor disk device according to the present invention is a semiconductor disk device having a volatile memory in which data is read / written in response to a disk access command, and has a capacity larger than that of the memory. A hard disk in which a plurality of different areas each having a capacity are set; and a save unit for storing data stored in the memory in the plurality of areas in response to power-off of the semiconductor disk device; A battery that supplies operating power to the evacuation unit in response to power cutoff.

With this configuration, when the power is turned off,
Data stored in the volatile memory is saved in each area provided for saving. Therefore, for example, even if there is a defect on the medium in one of the plurality of areas and the data is not saved normally, it is possible to normally save the data to another area. it can.

[0010] The semiconductor disk device further includes a restoring means for reading data stored in one of the plurality of areas and storing the data in the memory in response to the start of power supply, and A determination unit that determines whether or not an error has occurred in the data reading; and if the determination unit determines that an error has occurred, the data stored in an area different from the area where the error has occurred is stored in the memory. And a control means for controlling the restore means so as to store the data in the storage means.

According to such a configuration, even if there is a defect on the medium of the HDD, the data stored in another area is transferred to the memory, so that the data restoration processing at power-on can be performed. It can be executed normally and the destruction / loss of data stored in the semiconductor memory can be prevented, and the normal startup processing of the semiconductor disk device can be performed.

The restoring means reads data for each block of a predetermined size, and the control means, when the judging means judges that an error has occurred, the data for one block in which the error has occurred. The data may be read from an area different from the area where the error has occurred, and the restore unit may be controlled to store the data in the memory.

When the determination means determines that an error has occurred, the control means reads out data subsequent to the block in which the error has occurred from an area different from the area in which the error has occurred. The restore unit may be controlled to store the data in the memory. By configuring the control means in this way, effective data restoration processing can be realized even when each of the plurality of areas has a defect.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a semiconductor disk device according to the present invention will be described with reference to the drawings. FIG.
FIG. 1 is a block diagram showing a configuration when a semiconductor disk device according to a first embodiment of the present invention is applied to a computer system.

In the computer system 1 shown in FIG. 1, a semiconductor disk device 10 is connected to a host computer 30 via a bus 20. The semiconductor disk device 10 performs various processes according to commands from the host computer 30 sent via the bus 20. In this embodiment, the bus 20 is a SCSI (Small Computer).
System Interface) standards.

The semiconductor disk device 10 includes a control unit 11,
Memory 12, backup HDD (hard disk) 1
3, an external power supply (PS: Power Supply) having an interface 14, a switching circuit 15, a battery 16, a battery controller 17, and an internal bus 18.
It operates according to the power supplied from.

The control unit 11 controls the entire operation of the semiconductor disk device 10 and includes a microprocessor 11
0, F / W (firmware) memory 111, and registers Reg1 and Reg2. The microprocessor 110 is a central part of the control unit 11, and executes various processes such as a process according to a command sent from the host computer 30. The F / W memory 111 stores various programs necessary for the operation of the microprocessor 110. The registers Reg1 and Reg2 store a start address and a size of a save area described later, respectively. In this embodiment, when the power supply 19 is turned off, the register R
The contents stored in eg1 and Reg2 are saved in a predetermined area of the backup HDD 13.

The memory 12 is a volatile semiconductor memory such as a DRAM. For example, the memory 12 has a configuration in which there are eight SIMMs in which a 16-Mbyte semiconductor memory is set.
It has a memory capacity of M bytes. Backup HDD1
3 has a memory capacity at least twice the memory capacity of the memory 12, for example, a memory capacity of 340 Mbytes. Thus, the backup HDD 13 can store the data stored in the memory 12 in two different areas. In the following, these areas
As shown in (a) and (b), they are referred to as a save area 1 and a save area 2. Further, the backup HDD 13 stores
For example, a predetermined area of 512 bytes for storing data stored in the / W memory 111 and the registers Reg1 and Reg2 is provided in an area different from the save area 1 and the save area 2.

The interface 14 is an interface corresponding to the bus 20 conforming to the SCSI standard.
It interfaces with the host computer 30 and the like connected via the bus 20. Switching circuit 1
5 connects / disconnects the connection of the power supply line for transmitting the power supplied from the power supply 19 according to the control of the control unit 11. The power supply line for supplying the power transmitted from the switching circuit 15 is wired OR with the power supply line for supplying the power from the battery 16. An output line from the wired OR is connected to the microprocessor 110, the memory 12, and the backup HDD 13, and is configured to be supplied with power.

The battery 16 is realized by a lithium battery or the like, is built in the semiconductor disk device 10, and supplies power to a power supply line under the control of the battery controller 17. The battery controller 17 follows a command sent from the control unit 11 via the internal bus 18 to
The start / stop of power supply from the battery 16 is controlled.
However, during operation of the semiconductor disk device 10, power is constantly output from the battery 16.

The power supply 19 generates electric power necessary for the operation of the semiconductor disk device, and supplies the electric power according to an instruction from the switching circuit 15. In the first embodiment, the semiconductor disk device 10 is provided outside the device, but may be provided inside the device.

Here, the backup HD is stored in the memory 12.
FIG. 2 shows the process of saving and restoring data to D13.
A brief description will be given with reference to FIGS. As described above, the backup HDD 13 is provided with the save area 1 and the save area 2 having the same capacity as the memory 12. In the semiconductor disk device 10, when the power is turned off, the data stored in the memory 12 is stored in the backup area 1 of the backup HDD 13 as shown in FIG.
And the same data is saved in the save area 2 as well. On the other hand, when the power is turned on,
Restore processing for returning data from the backup HDD 13 to the memory 12 is performed. Here, as shown in FIG. 2B, when there is a defect on the medium in the save area 1 and the restoration processing of the data saved in the save area 1 is not completed (for example, the data When the reading has failed), the restoration processing is executed using the data saved in the save area 2.

Next, the processing of the control unit 11 when turning on / off the power of the semiconductor disk device will be described with reference to the flowcharts of FIGS. First, the data saving process when the power is turned off will be described. This power-off process is executed by a timer interrupt generated according to the count of a timer (not shown). The control unit 11 starts the process at the time of power shutdown in response to the generation of the interrupt signal. The control unit 11 first determines whether or not the power is shut off by detecting the state of the switching circuit 15, that is, whether or not the power is not supplied from the power supply 19 (step S <b> 1). Here, when the detected state of the switching circuit 15 does not indicate that the power is off,
The control unit 11 performs processing to detect the state of the switching circuit 15 again (steps S2 and S3). If the result of the re-detection does not indicate that the power is off, the control unit 11
Returns to the normal processing again. This indicates that some normal processing can be performed while the power of the semiconductor disk device 10 is turned on.

If the power-off is detected in step S1, or if the result of the re-detection indicates power-off (YES in step S3), the control unit 11
Starts the process of saving the data stored in the memory 12 to the save area 1 of the backup HDD 13. At this time, the control unit 11 executes a save process according to the start address and the size of the save area 1 stored in the register Reg1. When the power supply from the power supply 19 is interrupted, the power from the battery 16 is replaced with the power supplied from the power supply 19 instead of the power supplied from the power supply 19, as described above. Is supplied to

Next, the control section 11 determines whether or not the data saving has been completed normally (step S5). Here, the completion determination of the data saving is performed by the backup HDD 1
3 shows a determination as to whether or not a hardware mechanical problem has occurred, such as a head crash in which a data read / write head contacts a medium. If such a hardware problem occurs, the backup H
Data read / write processing for the DD 13 is not performed. That is, when the data saving processing is not completed normally (step S5, NO), a notification to the effect that the data saving processing cannot be performed normally is sent to the host computer 30. Then, processing for errors and abnormalities such as notifying the abnormality of the backup HDD 13 is performed (steps S6 and S7).

When the data saving to the save area 1 is completed normally (step S5, YES), the control unit 11 deletes the data stored in the memory 12, that is, the same data as the data saved to the save area 1. Is saved in the save area 2 (step S8). At this time, the control unit 11 executes the save processing with reference to the start address and the size of the save area 2 stored in the register Reg2. Thereafter, the control unit 11 determines whether or not the data saving has been completed normally as in step S5 (step S5).
9). If the data saving processing is not completed normally due to a head crash or the like, the processing for the data and the abnormality described above is performed (steps S6 and S7).

When the data saving to the saving area 2 is completed normally (step S9, YES), the control unit 11 sends an instruction to the battery controller 17 to stop the power supply from the battery 16.

According to the above processing, the data stored in the memory 12 is changed to the backup HD
It is saved in the save area 1 and the save area 2 of D13. still,
The above-described power cut-off process is executed in response to a timer interrupt by a timer. However, the power cut-off process is not limited to this. For example, an operation of turning off the switching circuit 15 by an operator of the computer system 1 may be performed. Alternatively, an interrupt signal may be generated when a command to shut off the power of the semiconductor disk device 10 is sent from the host computer 30 via the bus 20.

Next, data restoration processing at the time of power-on will be described with reference to FIG. First, a command to turn on the power is sent from the host computer 30 via the bus 20 or the computer system 1
When the operator performs the processing of turning on the switching circuit 15, the power for operation is supplied from the power supply 19 to each component of the semiconductor disk device 10, and an interrupt signal is generated. In response to the interrupt signal, the process at the time of turning on the power is started.

The control unit 11 first operates the backup HDD
13 is stored in the save area 1
2 (restore processing) is started (step S11). Thereafter, the control unit 11 determines whether data reading from the save area 1 has been completed (Step S12). That is, here, it is determined whether or not data has been correctly read from the save area 1 of the backup HDD 13. Here, if the data reading has been completed,
The control unit 11 controls the semiconductor disk device 1 in response to power-on.
Normal processing is started assuming that the startup of 0 has been completed.

If the reading of data from the save area 1 is not completed (step S12, NO), for example, there is a defect on the medium in the save area 1 as shown in FIG. Therefore, when the data transfer is not performed normally, the control unit 11
The process of transferring the data stored in the save area 2 to the memory 12 is started (step S13). After this,
The control unit 11 determines whether the data stored in the save area 2 has been normally read (Step S1).
4). Here, if the data reading has been completed,
The control unit 11 controls the semiconductor disk device 1 in response to power-on.
Normal processing is started assuming that the startup of 0 has been completed.

If the reading of data from the save area 2 has not been completed (step S14, NO), the control unit 11
Holds the status indicating that the restore process has failed in a predetermined register, and ends the process at the time of turning on the power (step S15).

After the restoration process has failed, some command is issued from the host computer 30 to the semiconductor disk device 1.
When the status is sent to 0, the status held in step S15 is read out, whereby the host computer 30 recognizes that the restore processing in the semiconductor disk device 10 has failed.

With the above processing, even if the data saved in one of the save areas is destroyed or lost due to a defect on the medium, the data can be restored according to the power-on using the data saved in the other save area. Processing can be performed.

Particularly, there is a progressive defect among the defects on the medium, and in the data saving process, the data is re-read (verification process), and it is recognized that the data is normally saved on the backup HDD medium. Even so,
Data may be destroyed or lost over time. Even in such a case, in the above-described embodiment, the data is saved in two different areas, and the data is destroyed or lost due to a progressive defect in one of the areas. Also, the restoration process can be executed using the data saved in the other area.

In the above-described first embodiment, two save areas are provided to save data stored in the memory. However, the present invention is not limited to this, and three or more save areas are different. It may be provided in a region.

Next, a second embodiment of the semiconductor disk device according to the present invention will be described. However, the configuration of the semiconductor disk device according to the second embodiment is the same as the configuration of the above-described first embodiment (see FIG. 1), and a detailed description thereof will be omitted. Further, the data saving process in the second embodiment is the same as that in the first embodiment, and a detailed description is omitted.

The second embodiment is characterized by data restoration processing, which will be described with reference to FIG. In the second embodiment, data restoration processing is performed for each block of a predetermined size, here, 512 bytes. The save area 1 and the save area 2 shown in FIG. 5 have the same capacity as the memory 12 like the save areas shown in FIGS. 2A and 2B.

Here, it is assumed that there is a media defect in the save area 1 as shown in FIG. In the second embodiment, when data reading is not performed normally due to the media defect, data is read from the save area 2 only for the block including the media defect.

Such a restore process will be described in detail with reference to the flowchart shown in FIG. Each component will be described using the same reference numerals as in the first embodiment.

First, the host computer 30 sends the bus 2
0, or when the operator of the computer system 1 performs a process of turning on the switching circuit 15, the power supply 19 sends an operation signal to each component of the semiconductor disk device 10. Power is supplied and an interrupt signal is generated. In response to the interrupt signal, the process at power-on is started.

The control unit 11 uses, for example, a variable n to
Execute restore processing while counting the number of blocks. Note that the memory 12 is composed of m blocks 1 to m. The control unit 11 first starts data transfer from the block 1 of the save area 1 to the memory (steps S21 and S22).

Here, the control section 11 determines whether or not the data reading of the block 1 has been normally performed (step S23). When the data reading is performed normally (step S23, YES), the control unit 11 determines whether or not the block number n is equal to the maximum block number m (step S24).

If the number of blocks n is equal to the maximum number of blocks m (step S24, YES), this restore processing ends. When the number of blocks is not equal to m (step S24, NO), n is incremented by one and the process returns to step S22 again (step S22).
25).

That is, unless data reading from the save area 1 fails, steps S22, S23, S24, S24
The restore processing is completed by the loop processing of 25.
If the data reading is not performed normally in step S23, the control unit 11 starts data transfer from the block n of the save area 2 to the memory (step S26). Thereafter, the control unit 11 determines whether or not the data reading of the block n has been normally performed (Step S27). If the data reading has been performed normally (step S27, YES), the process returns to step S24.

When the data reading of the block n is not normally performed from the save area 2 (step S27,
NO), the control unit 11 holds a status indicating that the restoration process has failed in a predetermined register, and ends the process at the time of power-on (step S28).

After the restoration process has failed, some command is issued from the host computer 30 to the semiconductor disk device 1.
When the status is sent to "0", the status held in step S28 is read out, whereby the host computer 30 recognizes that the restore processing in the semiconductor disk device 10 has failed.

By the above processing, even if the data saved in one of the save areas is destroyed or lost due to a defect on the medium, the data is restored in response to power-on using the data saved in the other save area. Processing can be performed.

In particular, in the second embodiment, data is read from another save area only for a block where a defect is detected, so that effective restore processing is realized even if each save area has a defect. can do.

Next, a third embodiment of the semiconductor disk device according to the present invention will be described. However, the configuration of the semiconductor disk device according to the third embodiment is the same as the configuration of the above-described first embodiment (see FIG. 1), and a detailed description thereof will be omitted. Further, the data saving process in the third embodiment is the same as that in the first embodiment, and the detailed description is omitted.

The third embodiment is also characterized by the restore processing, as in the second embodiment. Particularly, in the third embodiment, unlike the above-described second embodiment, when data cannot be read due to a defect in the first save area, the subsequent data read is performed from the second save area. When the data cannot be normally read in the second save area, the data is read again from the first save area.

The details of the restore process will be described with reference to the flowchart shown in FIG. Also in the third embodiment, the control unit 11 executes the restoration process while counting the number of blocks using the variable n. Similarly, the memory 12 is composed of m blocks 1 to m.

First, the host computer 30 sends the bus 2
0, or when the operator of the computer system 1 performs a process of turning on the switching circuit 15, the power supply 19 sends an operation signal to each component of the semiconductor disk device 10. Power is supplied and an interrupt signal is generated. In response to the interrupt signal, the process at power-on is started.

The control unit 11 starts data transfer from the block 1 of the save area 1 to the memory (step S).
31, 32). Here, the control unit 11 controls the block 1
It is determined whether or not the data has been read normally (step S33). When the data reading is performed normally (step S33, YES), the control unit 11 determines whether or not the block number n is equal to the maximum block number m (step S34).

If the number n of blocks is equal to the maximum number m of blocks (step S34, YES), this restore processing ends. If the number of blocks is not equal to m (step S34, NO), n is incremented by one and the process returns to step S32 again (step S34).
35).

That is, unless data reading from the save area 1 fails, steps S32, S33, S34, S34
The restore process is completed by the loop process of 35.
If the data reading is not performed normally in step S33, the control unit 11 starts data transfer from the block n of the save area 2 to the memory (step S36). Thereafter, the control unit 11 determines whether the data reading of the block n has been normally performed (step S37). If the data reading has been performed normally (step S37, YES), the control unit 11 determines whether the number n of blocks is equal to the maximum number m of blocks (step S38).

If the number n of blocks is equal to the maximum number m of blocks (step S38, YES), the restore processing ends. If the number of blocks is not equal to m (NO in step S38), the control unit 11 starts data transfer from the block n in the save area 2 to the memory after incrementing n by one (step S39, NO). S
40).

Thereafter, the control section 11 determines whether or not the data reading of the block n has been normally performed (step S41). When the data reading is performed normally (step S41, YES), the control unit 11 determines whether or not the block number n is equal to the maximum block number m (step S42).

When the number of blocks n is equal to the maximum number of blocks m (step S42, YES), the restore processing ends. If the number of blocks is not equal to m (step S42, NO), the controller 11
It returns to the process of 39.

That is, once data is read from the save area 2, unless the data read fails,
Data reading from the save area 2 is continued (steps S39, S40, S41, S42).

In the step S41, when the data reading is not performed normally (step S41, N
O), the control unit 11 starts data transfer from the block n of the save area 1 to the memory (step S43). Thereafter, the control unit 11 determines whether the data reading of the block n has been normally performed (step S44).
When data reading is performed normally (step S4
4, YES), the control unit 11 returns to the process of step S34 again.

That is, when data is read from the save area 1 again, unless the data read fails,
Data reading from the save area 1 is continued (steps S34, S35, S32, S33).

If it is determined in steps S37 and S44 that the data has not been read normally, it means that the data has not been normally read from the save area 1 and the save area 2 for the same block. It means that it was not issued. Therefore, in this case (step S37, NO or step S44, NO), the control unit 1
1 holds a status indicating that the restoration process has failed in a predetermined register, and ends the process at the time of power-on (steps S45 and S46).

After the restoration process has failed, some command is issued from the host computer 30 to the semiconductor disk device 1.
0, the status held in steps S45 and S46 is read out, whereby
The host computer 30 recognizes that the restore processing has failed in the semiconductor disk device 10.

With the above processing, even if the data saved in one of the save areas is destroyed or lost due to a defect on the medium, the data can be restored according to the power-on using the data saved in the other save area. Processing can be performed.

In particular, in the third embodiment, data is read from another save area only for a block in which a defect is detected, and therefore, as in the second embodiment,
Even when each save area has a defect, effective restore processing can be realized.

[0067]

As described above in detail, according to the present invention, a plurality of HDD save areas for backing up data stored in a volatile memory are provided. Save the data stored in. Thus, even if there is a defect on the medium of the HDD, the data stored in another area is transferred to the memory, so that the data restoration process at the time of turning on the power can be normally executed, and the semiconductor memory is restored. Destruction and loss of stored data can be prevented, and normal startup processing can be performed.

[Brief description of the drawings]

FIG. 1 is an exemplary block diagram showing a configuration of a computer system to which a semiconductor disk device according to a first embodiment of the present invention is applied;

FIG. 2 is a conceptual diagram for explaining a data saving process from a volatile memory to a backup hard disk and a restore process from a hard disk to a memory in the first embodiment.

FIG. 3 is a flowchart for explaining data saving processing from a volatile memory to a backup hard disk in the first embodiment.

FIG. 4 is a flowchart illustrating a process of restoring data from a backup hard disk to a volatile memory in the first embodiment.

FIG. 5 is a conceptual diagram illustrating a process of saving data from a volatile memory to a backup hard disk and a process of restoring data from a hard disk to a memory in the second embodiment of the present invention.

FIG. 6 is a flowchart illustrating a process of restoring data from a backup hard disk to a volatile memory in the second embodiment.

FIG. 7 is a flowchart illustrating a process of restoring data from a backup hard disk to a volatile memory according to the third embodiment of the present invention.

FIG. 8 is a conceptual diagram illustrating a conventional data save process from a volatile memory to a backup hard disk and a restore process from a hard disk to a memory.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Computer system, 10 ... Semiconductor disk device, 11 ... Control part, 12 ... Memory, 13 ... Backup HDD, 14 ... Interface, 15 ... Switching circuit, 16 ... Battery, 17 ... Battery controller,
18 internal bus, 19 power supply (PS), 20 bus, 3
0: Host computer.

Claims (4)

[Claims] 1. A semiconductor disk device having a volatile memory from / to which data is read / written in response to a disk access command, comprising: a hard disk in which a plurality of different areas each having a capacity larger than the capacity of the memory are set; An evacuation unit for storing data stored in the memory in the plurality of areas in response to the interruption of the power supply of the semiconductor disk device; and supplying operating power to the evacuation unit in response to the interruption of the power supply. A semiconductor disk device comprising a battery. 2. Reading data stored in one of the plurality of areas according to the start of power supply,
Restoring means for storing the data in the memory; determining means for determining whether an error has occurred in reading data by the restoring means; area where the error has occurred if the determining means determines that an error has occurred; 2. The semiconductor disk device according to claim 1, further comprising control means for controlling said restore means so as to store data stored in an area different from said area in said memory. 3. The restoration unit reads data for each block of a predetermined size. When the determination unit determines that an error has occurred, the control unit determines whether the error has occurred in one block. 3. The semiconductor disk device according to claim 2, wherein said restoring means is controlled so that said data is read from an area different from the area where said error has occurred and stored in said memory. 4. The restoration means reads data for each block of a predetermined size. When the determination means determines that an error has occurred, the control means reads the data after the block in which the error occurred. 3. The semiconductor disk device according to claim 2, wherein the restore unit is controlled to read data from an area different from the area where the error has occurred and store the data in the memory.