JP2960455B2 - External storage control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] In a disk cache system, an external storage control method for controlling write-back of data in an auxiliary memory that temporarily holds write data is described. A central processing unit (11), an external storage unit (12), a cache memory (14), and the central processing unit (10) are provided for controlling whether or not writing of interrupted data to an external storage device can be controlled. An auxiliary memory (13) for storing write data from the central processing unit (11) to the cache memory (14); and writing the write data to the cache memory (14) when there is a write from the central processing unit (11) to the cache memory (14). (14) and also stored in the auxiliary memory (13). At an appropriate timing, the auxiliary memory (13) is controlled. Wherein by writing the write data on the external storage device (12) a central processing unit (11) and said external storage device (1
2) Disk controller for high-speed writing (15)
An external storage control method for an external storage control system, comprising: the disk controller (15), when the write processing to the auxiliary memory (13) is interrupted, notifies the central processing unit of the reason for the interruption of the write processing. (11). The central processing unit (11) sends the disk controller (15) to the external storage device (12) based on the interruption reason notified by the disk controller (15). The processing to be performed on the disk controller (1
5), wherein the disk controller (15) keeps the write processing stored in the auxiliary memory (13) interrupted in accordance with the contents of the good intention processing instructed by the central processing unit (11). It is configured to select whether to write the write data to the external storage device (12).

[Industrial applications]

The present invention relates to an external storage control method for controlling writing of data held on an auxiliary memory for temporarily holding write data to an external storage device in a disk cache system, and in particular, from a central processing unit (hereinafter referred to as a CPU). A disk controller that temporarily stores write data for the cache memory in the auxiliary memory, transfers the write data on the auxiliary memory to the external storage device at an appropriate timing and writes the data, thereby performing high-speed writing between the central processing unit and the external storage device. The present invention relates to an external storage control method of an external storage control system provided.

[Conventional technology]

In a disk cache system, a cache memory having a higher speed and a smaller capacity than an external storage device is provided between a CPU and an external storage device, and at the time of reading, data read from the external storage device is temporarily stored in a cache. For example, when there is no other processing), the data in the cache memory is read. At the time of writing, the write data is temporarily stored in a cache memory, and the write data is transferred to an external storage device and written at an appropriate timing.

This allows the CPU to write and read to and from the external storage device at a high processing speed similar to that for a normal memory, without being aware of the difference in processing speed from the external storage device or the current operation state of the external storage device. It can be carried out.

In particular, when the external storage device is a magnetic disk device, it takes a long time to access the disk due to the rotation of the magnetic disk and the like. However, such a disadvantage can be solved by employing a disk cache system.

In this disk cache system, even if reading fails due to an error or power down when reading data from the external storage device, it is possible to directly access the external storage device and read the desired data as necessary. There is no inconvenience that data is lost.

However, in the case of data writing, when writing data in the cache memory to an external storage device and the data writing fails due to power down, instantaneous interruption, error, etc., the CPU writes the data to the memory. Since the end of processing is purging the write data, DKC
Cannot report the error even if the error is reported, and as a result, a problem arises that the write data is lost.

In order to prevent such a risk of data loss occurring at the time of data writing and to preserve data, a method of backing up the cache memory with a battery so that the data written to the cache memory is not lost due to a power down or the like has been considered. . However, this data preservation method of backing up the entire cache memory with a battery is actually implemented when the capacity of the cache memory becomes large, such as several hundred megabytes, as in a large system, since the backup battery becomes large in capacity. It is difficult.

In order to cope with this problem, a battery-backed non-volatile auxiliary memory having a smaller capacity is provided separately from the cache memory, the write data is stored in the non-volatile memory in parallel with the cache memory, and There has been proposed a data security system in which management information of write data is written in a nonvolatile memory.

By doing so, even if the cache memory has a large capacity, it is possible to reliably maintain write data with a small-capacity nonvolatile memory and a small backup battery.

[Problems to be solved by the invention]

As described above, the disk cache system can reliably maintain write data by storing the write data and management information in a small-capacity nonvolatile memory provided separately from the cache memory.

However, if an error occurs in the write data due to an instantaneous interruption of the power supply, an error, or the like while writing data to the non-volatile memory in order to write to the external storage device, the error data is written in the subsequent write back. There is a problem that the data format on the medium of the external storage device is destroyed by writing to the external storage device.

According to the present invention, when an interruption occurs in the writing process to the non-volatile memory due to an error or the like, the host device (CP
An object of the present invention is to provide an external storage control method of a disk controller which is made controllable by U) and improved so as to prevent data in an external storage device from being destroyed by error data.

[Means for solving the problem]

When an error or the like occurs during writing to the memory in the disk cache system, the writing of the interrupted data on the memory is executed as it is, and the data portion of the external storage device may be destroyed. In some cases, it may be necessary to stop writing to save the data portion of the external storage device. For example, the top CP
If the data writing is temporarily terminated due to U, the write-back may be temporarily stopped, the data before the writing may be saved, and the failed process may be resumed for recovery.

Thus, when writing to memory is interrupted,
The way to respond depends on the reason for interruption. The present invention pays attention to this point, and when the writing to the memory is interrupted, it is possible to control whether or not to execute the write-back depending on the interruption reason.

Hereinafter, means employed by the present invention to solve the above-described problem will be described with reference to FIG. FIG. 1 is a block diagram showing the configuration of an implementation system of the present invention.

In FIG. 1, reference numeral 11 denotes a CPU, which activates an input / output command for a desired external storage device and performs processing for transferring data for writing / reading to a predetermined external storage device.

Reference numeral 12 denotes an external storage device such as a magnetic disk device, which stores data to be written / read and management information thereof.

An auxiliary memory 13 stores data to be written in the cache memory in parallel and also writes necessary management information. The auxiliary memory is a nonvolatile memory backed up by a battery.

Reference numeral 14 denotes a cache memory for temporarily storing write / read data for the external storage device 12.

Reference numeral 15 denotes a disk controller (hereinafter referred to as DKC), which receives input / output instructions from the host CPU 11 and temporarily stores write / read data in the cache memory 14 by a disk cache method, and requires write data. It also controls operations such as a process of writing to the auxiliary memory 13 together with various management information and a process of transferring data between the external storage device 12 and the cache memory 14.

The present invention is implemented by the external storage control system described above, and is configured as follows. That is, a central processing unit (11), an external storage device (12), a cache memory (14), and an auxiliary memory (13) for storing write data from the central processing unit (11) to the cache memory (14). And when the central processing unit (11) writes to the cache memory (14), the write data is stored in the cache memory (14) and also stored in the auxiliary memory (13). The central processing unit (11) and the external storage device (1) write the write data on the auxiliary memory (13) to the external storage device (12) at an appropriate timing.
2) Disk controller for high-speed writing (15)
An external storage control method for an external storage control system comprising: the disk controller (15), when the write processing to the auxiliary memory (13) is interrupted, notifies the central processing unit of the reason for the interruption of the write processing. (11). The central processing unit (11) sends the disk controller (15) to the external storage device (12) based on the interruption reason notified by the disk controller (15). The processing to be performed on the disk controller (1
5) wherein the disk controller (15) keeps the writing process stored in the auxiliary memory (13) interrupted in accordance with the processing content instructed by the central processing unit (11). Selecting whether to write the write data to the external storage device (12).

(Operation)

When data is written from the CPU 11, the DKC 15 writes the write data to the cache memory 14. At the same time, the write data is stored in parallel with the necessary management information in the auxiliary memory 13.

When the data writing to the cache memory 14 and the auxiliary memory 13 is completed, the DKC 15 performs a process of transferring the write data on the auxiliary memory 13 to the external storage device 12 at an appropriate timing.

If it is detected that the writing was interrupted due to an instantaneous power interruption or an error while writing to the memory (14, 13), the DKC 15 interrupts the writing process and informs the CPU 11 of the reason for the interruption of the writing process. Notice.

The central processing unit 11 instructs the DKC 15 on processing contents such as whether or not to perform the writing processing on the external storage device 12 based on the interruption reason. For example, when the history of old data is necessary to restore data, an instruction to stop writing to the external storage device 12 is issued, and when invalidating the write data portion, writing to the external storage device 12 is performed. To instruct.

According to the instruction from the CPU 11, the DKC 15 selects a processing mode such as executing the writing of the interrupted data on the memory to the external storage device 12, stopping the writing, and saving the old data of the external storage device 12, and the like. .

As described above, even if writing is interrupted during memory write, the upper CPU receives notification of the reason for the write interruption and instructs DKC to respond. Control of the processing mode such as whether or not the write data can be performed, thereby enabling appropriate processing corresponding to each interruption reason, making it possible to reliably determine the validity of the write data, and erroneously outputting error data to the external device. Data in the storage device can be prevented from being destroyed.

〔Example〕

One embodiment of the present invention will be described with reference to FIGS. FIG. 2 is an explanatory diagram of a track format of an external storage device, FIG.
FIG. 4 is an explanatory diagram of a directory on the auxiliary memory used in one embodiment of the present invention, and FIG. 5 is an explanatory diagram of a data format on the auxiliary memory used in the embodiment.

Since the data read operation of the external storage device 12 using the cache memory 14 is the same as that of the conventional method, the following description will focus on the data write operation.
12 is D for DASDCPU data processing unit such as magnetic disk drive
A case will be described as an example in which the unit is an ASD track.

(A) Configuration of the implementation system In FIG. 2, the CPU 11, the external storage device 12, the auxiliary memory 13, the cache memory 14, and the DKC 15
As described in the figure.

In the auxiliary memory 13, 131 is a directory,
As described in FIG. 4, processing information for each processing unit is written. Reference numeral 132 denotes a data area in which write data is written in a record format, and as shown in FIG. 5, record information consisting of various signals indicating the state of each field of the record is recorded in a record descriptor 133.
Is written to.

In the DKC 15, reference numeral 151 denotes a status register, in which various flags indicating a reason for interruption of writing are set.

FIG. 3 shows a track format when data is written to the external storage device (DASD) 15. The track format starts with the index and ends with the index.

HA is a home address, and indicates a physical position of the track and a normal or abnormal state of the track.

R0 is the first record following the home address HA,
It is used to indicate a replacement track when this track is defective.

R1 to RN following R0 are records to which data is written. Each record is composed of three areas of a count area C, a key area K, and a data area D, as shown in the record R1 in an enlarged manner. Is done.

In the count area C, the number of the record in the track,
The key length and the data length are written, from which the size of the record can be known. In the key area K, a physical index for searching a record is written.

In the data area D, user data including data information defined by the count area C and the key area K is written in segments (32 bytes).

Since the number of records R1 to RN is variable, the data written in this track format is variable length data.

FIG. 4 shows the structure of the directory 131 in the auxiliary memory 13. The sequence number of the directory is written in the sequence number area. Each time the directory information is updated, the sequence number is also updated.

In the start / end area, a start position s and an end position e of a record storing data are written as memory addresses on the auxiliary memory 13. In the block address area, an address of a block in which data is stored is written as a memory address on the auxiliary memory 3.
Each data in the track is divided into a plurality of blocks, and each block is dispersedly arranged in the memory in order to increase the efficiency of use of the memory, but is logically one block.

The directory information written as above is
It relatively corresponds to the directory information on the cache memory 15.

FIG. 5 shows the format of data to be written on the auxiliary memory 13, and includes records R1 and R2.
Are exemplified. Count area C ₁ , key area
K ₁ and data area D ₁ represents a data record R1, count area C _2, key area K ₂ and data area D ₂ record R2
The data of is shown. There is no cap between the areas as in the DASD12 track format.

RD (RD ₁ , RD _2, etc.) is a record descriptor 133, which consists of 2 bytes, is provided after the count area C, and record information as shown is written.

That is, the flag 1 is set to the next count-in progress flag when the count area C is being updated. The key-in progress flag is set to 1 when the key area K is being updated. The data-in-progress flag is set to 1 when the key area K is being updated.

The next count updated flag is set to 1 when the count area C has been updated. The key updated flag is set to 1 when the key area K has been updated.
Is set. The data updated flag is set to 1 when the key area K has been updated.

The format write flag is a flag indicating whether or not the data format has been updated, and is set to 1 when the format is newly rewritten as in a write CKD command.

The end process flag is a flag indicating the temporary termination of the processing, and is set to 1 when the data writing processing is temporarily terminated.
Is set.

As described above, by writing the record information in the record, the processing state of each area can be specified.

For example, during the update of the data of the count area C ₁ of the record R2 record descriptor RD ₁ of the previous record R1
The next count-in progress flag is set to 1 and the next count-in progress flag is reset and the next count updated flag is set to 1 when the updating process of the count part of the record R2 is completed. I do.

As a result, the record descriptor RD of the record R1 is obtained.
₁ next count in progress flag is 1
In, when the next count Apu Day Ted flag is 0, it can be detected that the processing for some reason during the data update processing of the count area C ₁ of the record R2 is interrupted. Also, the next count of record descriptor RD ₁
When the updated flag is 1, it is detected that the updating process of the count portion of the record R2 has been completed normally. The same holds true for updating the respective data key area K ₂ and data area D _2.

(B) Operation of Embodiment The write operation of the embodiment will be described in accordance with the processing sequence thereof with reference to FIGS. 2 to 5, taking a write operation by a write CKD as a general write operation as an example. The writing of data to the auxiliary memory 13 is performed in parallel with the writing to the cache memory 14. However, the writing to the cache memory 14 is performed in the same manner as in the conventional method, and the description of the writing operation is omitted.

When a write CKD command is issued from the CPU 11 and the write data is transferred, the DKC 15 writes the write data to the cache memory 14 and stores the write data in the auxiliary memory 13 in parallel.

Writing of data to the cache memory 14 and the auxiliary memory 13 is performed in units of records according to the track format shown in FIG. 3, and each record is performed in the order of the count area C, the key area K, and the data area D.

Further, at the same time as writing the record data to the auxiliary memory 13, directory information, which is processing information for each processing unit (track unit), is written to the directory 131 as described with reference to FIG. In the count area C of each record, as described with reference to FIG. 5, record information indicating the state of each field (CKD area) of the record is written to its record descriptor RD.

When the writing of data to the cache memory 14 and the auxiliary memory 13 is completed, the DKC 15 writes the write data on the auxiliary memory 13 to the external storage device 12 at an appropriate timing without any other processing in the same manner as in the conventional disk cache system. Perform the transfer process.

If the writing is interrupted due to an error of the CPU or DKC at any point during the writing of the data, the DKC stops updating the memory.

When the above processing is completed, the DKC 15 reports the contents of the status register to the CPU.

By checking the contents of the status register 151, the CPU 11 detects the reason for the interruption of the writing process.

The CPU 11 determines the content of processing such as whether or not to perform this write processing on the external storage device 12 based on the detected interruption reason.
Instruct DKC15.

For example, when the history of the old data is needed to restore the data, a cancel command for instructing to stop writing to the external storage device 12 is issued.

To invalidate the write data part, use the external storage device.
A write command instructing writing to 12 is issued.

The DKC 15 executes each process according to the instruction from the CPU 11.

In the case of a cancel command, write-back is performed after erasing the data of the interrupted process on the auxiliary memory 13. In the case of a write command, the write-back is performed to the external storage device 12 while the data having the error is stored, and the write is performed at a predetermined address position indicated by the directory 131.

Although one embodiment of the present invention has been described above, the reasons for interrupting writing according to the present invention are not limited to the embodiment, and various reasons for interruption can be set in accordance with the gist of the invention.

〔The invention's effect〕

As described above, according to the present invention, the following effects can be obtained.

(1) Even if writing is interrupted during a memory write, the upper CPU receives notification of the reason for the write interruption and responds with DKC
Therefore, it is possible to control the processing mode such as whether or not writing can be performed in response to each interruption reason, and it is possible to perform appropriate recovery processing corresponding to each interruption reason.

(2) According to the above (1), the validity of the write data can be reliably determined, and it is possible to prevent data in an external storage device that does not require error data from being destroyed.

[Brief description of the drawings]

FIG. 1 is an explanatory view of an embodiment of the present invention, FIG. 2 is an explanatory view of an embodiment of an embodiment of the present invention, FIG. 3 is an explanatory view of a track format of an external storage device, and FIG. FIG. 5 is an explanatory diagram of a directory on an auxiliary memory used in one embodiment. FIG. 5 is an explanatory diagram of a data format on an auxiliary memory used in the embodiment. In FIG. 1 and FIG. 2, 11: central processing unit (CPU), 12: external storage device, 13
... Auxiliary memory, 131 directory, 132 data area, 133 record descriptor, 14 cache memory, 15 disk controller (DKC).

Claims (1)

(57) [Claims] A central processing unit (11) and an external storage device (1)
2) a cache memory (14); an auxiliary memory (13) for storing write data from the central processing unit (11) to the cache memory (14); and a cache memory from the central processing unit (11). When there is a write to (14), control is performed so that write data is stored in the cache memory (14) and also stored in the auxiliary memory (13). The write data to the external storage device (12)
An external storage control system of an external storage control system including a disk controller (15) for performing high-speed writing between the central processing unit (11) and the external storage device (12) by writing to the disk controller; (15) Notifying the central processing unit (11) of the interruption reason of the writing process when the writing process to the auxiliary memory (13) is interrupted, wherein the central processing unit (11) Based on the interruption reason notified by the disk controller (15), the disk controller (15) determines the processing contents to be performed on the external storage device (12) by the disk controller (15).
The disk controller (15) communicates with the central processing unit (1).
According to the content of the processing instructed from 1), it is to select whether or not to write the write data stored in the auxiliary memory (13) while the write processing is interrupted to the external storage device (12). An external storage control method characterized by the following.