JP3226047B2 - I / O device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention protects data by performing dual management of data with different disk volumes,
Are those concerning the input and output device to improve the reliability of the system, it is to allow rapid reliable recovery of a volume, especially when the one disk failure during the update of the data.

[0002]

Background of the Invention and out to perform a conventional dual volume processing
In the input device, an I / O command (SIO) is issued to each volume of two disks to write the same data by the function of the host data management system (IOS), and even if one disk fails, The business can be continued by the volume of the other disk. At the time of writing this data, update information such as the number of updates and the number of the updated cylinder is recorded so that the last updated cylinder is always known. The update information is recorded by providing a specific track on the disk or providing a nonvolatile semiconductor memory.

FIG. 5 shows such a dual volume processing.
1 shows a configuration example of a conventional input / output device for performing the above.
In the figure, 1 is a host, 2 is a disk controller, 3 and 4 are disk devices, and 5 and 6 are disk update information.

When updating data, the host 1 issues an SIO to each of the disk devices 3 and 4 and causes the disk control device 2 to write the same data. After updating the data for each disk, the host 1 records update information such as the number of updates and the write cylinder number.

[0005] Thereby, the update status of the two disks is kept the same. If one of the updates fails, the update information of the two disks is different. Therefore, if the cylinder is copied from the disk with the larger number of updates to the disk with the smaller number of updates, the equivalence can be guaranteed.

The conventional method is effective when the area to be updated by one SIO is limited to one cylinder. For example, the CKD command set which cannot execute writing beyond a cylinder boundary is impossible. (In the CKD command set, in the case of writing beyond a cylinder boundary, an instruction for positioning to the next cylinder must be given by issuing a CCW (channel command word)). No).

However, in the CKD-EX command set, which is an extended channel command body, WRITE UP
When several DATE DATA (data update write) commands are chained, it is allowed to update data beyond one cylinder.

For this reason, in the above-mentioned CKD command set, the CCW issuer can manage the last write cylinder, but in the CKD-EX command set, it cannot manage the last write cylinder. If the equivalence of two disk volumes is lost, either copy the entire disk or use extents (a physically separated storage area) for recovery.
Data had to be copied to the unit.

As a result, there is a problem that the amount of data copy increases and it takes time to recover.

[0010]

SUMMARY OF THE INVENTION According to the present invention, the host
It is an object of the present invention to provide means for making it possible to acquire the last write cylinder as update information in a dual volume processing system in which the last write cylinder cannot be specified when an IO is issued.

[0011]

SUMMARY OF THE INVENTION The present invention solves the above-mentioned problems by providing a mechanism for detecting cylinder switching and recording it as update information in a disk control device for controlling a disk device for constructing a duplex volume. It is intended.

FIG. 1 is a block diagram showing the principle of the present invention shown in an exemplary manner. In FIG. 1, reference numeral 1 denotes a host that performs processing using a duplicate volume.

Reference numeral 2 denotes a disk control device. 3, 4
Is a disk device on which a duplicate volume is constructed. These disk devices are hereinafter referred to as device X,
It is represented by device Y.

Reference numerals 5 and 6 denote update information of the respective disk devices of the device X and the device Y, and are stored in, for example, the nonvolatile semiconductor memory of the disk control device. Update information
The number of updates and the addresses of the start cylinder and end cylinder at the time of the last update are included.

Reference numeral 7 denotes a cylinder switching detection mechanism, which is provided in the disk controller 2 and detects when a cylinder switching has occurred in continuous data writing to the device X or the device Y.

The update information 5 and 6 are basically set and updated in accordance with an instruction from the host 1, but a part is instructed by the cylinder switching detecting mechanism 7, and when the cylinder switching is detected, the final cylinder address is updated. Is updated.
It should be noted that the host and the disk control device may be single or plural, respectively.

[0017]

In FIG. 1, a host 1 manages data by duplicating a volume by using disk devices of a device X and a device Y, and when updating data, the host 1 SIO is issued, a channel command or a command chain is started, and data is written to each device. At this time, the host notifies the disk controller 2 of the number of data updates and the address of the update start cylinder. The disk controller is
When data writing to the device is completed, the number of updates,
The update information including the update start cylinder and the update end cylinder is updated.

As described above, the channel command system is 1
If only one cylinder is written for one SIO issuance, the host knows the address of the cylinder to be updated, so the update information always indicates the last updated cylinder of the device. In the case where is permitted to write beyond the cylinder boundary for one SIO issuance, the host cannot know the final cylinder only by knowing the write start cylinder. However, in the present invention, the write beyond the cylinder boundary is detected by the cylinder switching detection mechanism of the disk controller, and the address of the last cylinder in the update information is updated based on the detection result. It is possible to correctly indicate the last cylinder.

For example, 100 disks are used for the disks of device X and device Y constituting one duplex volume.
It is assumed that the system is powered off the first time, the writing to the device X is completed first, and then the writing to the device Y is performed. At this time, it is assumed that the command chain activated by the SIO issued to the device X is as follows.

DX-LR (CCHH = 02.0D) -WU
D-WUD-WUD-WUD Here, DX is a command for instructing a seek, and LR (CCH
H = 0.2.0D) is a command for instructing a start address (cylinder 2, record 0D), and WUD is a command for instructing data update writing (Write Update Data).
WUD writes in record units, in this case 16
Assuming record / one track, cylinder switching occurs at the third WUD, and the cylinder address is changed from 2 to 3
Turns into. When updating the device X, the cylinder switching detection mechanism 7 detects this cylinder switching by reading the index mark appearing next to the last record on the track, and updates the update information 5 with the update count and the update number after the update is completed. The start address and end address of the cylinder are recorded. When updating the device Y, if the power failure occurs before the end of the cylinder update, the update information 6 is not rewritten.

As a result, the update information 5 of the device X is "1
00th time, cylinders 2 to 3 ", but the update information 6 of device Y remains the information at the time of the last update, for example," 99th time, cylinders 4 to 4 ".

When the system is started up after the restoration of the power failure, the update information 5 and 6 of the devices X and Y in which the duplicated volumes are defined are read and compared. In this case, it is detected that the equivalence of the update information of both the X and Y systems has been lost due to the difference in the number of updates, and by analyzing the difference, the 100th cycle of cylinders 2 to 3 of device X is performed. It can be understood that the write data of the above should be copied to the device Y.
This copy restores the equivalence.

[0023]

FIG. 2 is a block diagram of an input / output device for performing a dual volume processing according to an embodiment of the present invention. FIG. 2 (a) shows a host 1, a disk control device 2, and a device "10".
0 ”, disk devices 3 and 4 identified by device“ 200 ”, update information 5 and 6, cylinder switching detection mechanism 7
Fig. 2 (b) shows three hosts 1,
1 ', 1 ", two disk controllers 2, 2', at least four disk devices 3, 3 ', 4, 4', update information 5, 6, and a cylinder switching detection mechanism 7 The system shown in FIG. 2A corresponds to the principle configuration shown in FIG. 1 and shows the minimum configuration of an input / output device for performing a duplex volume process. Has a more complicated configuration than the system of (a), and includes a disk controller 2 for controlling the disk devices 3 and 3 'identified by the devices "100" and "101", and a device "200" and "201". Disk unit 4 to be identified
A disk controller 2 'for controlling the disk controller 4', three hosts 1, 1 ', 1 "are connected to the disk controllers 2, 2', respectively. 7, 7 ', and manages the update information 5, 6, and the device "100" and the device "20".
0, or device “101” and device “20”
A duplicate volume is constructed between 1 "and the like.
An embodiment will be described with a focus on the system of FIG.

FIG. 3 is a block diagram showing an embodiment of the inside of the disk control device 2 of FIG. 2 (b). Note that the configuration of the disk control device 2 'in FIG. 2B is substantially the same, and a description thereof will be omitted. 3, reference numeral 2 denotes a disk controller, 5 denotes update information, 7 denotes a cylinder switching detection mechanism, 8 denotes a non-volatile memory that stores the update information 5, 9 denotes a command analyzer that analyzes channel commands, and 10 denotes the update information 5. Is an update information management unit that creates Update information 5 is created for each host,
Is managed.

Next, the operation of the embodiment shown in FIGS. 2 (b) and 3 will be described using the example of the update information shown in FIGS. 4 (a) and 4 (b). (1) Each host 1, 1 ', 1 "declares use of a duplicated volume when the system is started up, and requests the disk control devices 2, 2' for a host identification number. (2) Each disk The control devices 2 and 2 'include hosts 1, 1',
ID numbers 00 and 0 in response to requests from
Assign 1,02. (3) The host writes SIO to the subsequent devices.
, The following two commands are chained at the head of the CCW to issue the CCW.

[0027] SELECT HOST BUFFE
R Notify the disk controller of the host identification number assigned to the host.

WRITE UPDATE INFO
RMATION Requests that the update information created by the host be transferred and stored in the disk controller. The update information includes the number of updates indicating the number of times of writing and the head address of the cylinder to be updated by the SIO. (4) When the command analysis unit 9 in FIG. 3 identifies these commands, it causes the update information management unit 10 to create the update information 5 corresponding to the host identification number and store it in the nonvolatile memory 8. (5) Here, the update information 5 in the disk controllers 2 and 2 '
6 is as shown in FIG. 4A, where the host "00"
In this state, the device “100” and the device “200” have been written 45 times, and the host “01” has been written in the device “100” and device “200” 60 times. Also, device “100”
The start address and the end address of the last updated cylinder of the duplicated volume match between the device and the device “200”. (6) After this, there is a 46th write from the host “00”, and there is a 61st write from the host “01”, and when the update to the device “100” has been completed, the system has lost power. Shall be. (7) The 46th write from the host “00” is 45
Starting from the cylinder “004” following the last cylinder “003” at the time of the first writing, the writing is performed on the next cylinder “005” beyond the cylinder boundary. When this cylinder boundary is exceeded, it is detected by the cylinder switching detection mechanism 7 of FIG. 3, and the update information management unit 10 updates the end address of the update cylinder of the update information 5. On the other hand, the 61st write from the host “01” is performed for the same start cylinder “006” to end cylinder “008” as in the case of the 60th write. Since these writings are performed only on the device "100", the update information 5 is updated as shown in FIG. 4B, but the update information 6 is the same as that in FIG. 4A. Therefore, the equivalence between the update information 5 and the update information 6 is lost. (8) In this state, if the power outage is restored and the system is started up, the disk controllers 2 and
Update information 5 and 6 are read from 2 'and compared to check for equivalence. From the result, it is found that the writing status of the device “100” is different from the writing status of the device “200”. Therefore, the device “100” is changed to the device “200”, the host “00” is changed from the cylinder “004” to “005”, “01” is the cylinder “006” to “00”
8 ". The contents of the update information 6 are also updated together with this copy, and the equivalence with the update information 5 is guaranteed.

The disk controller can create the update count for each device from the write control information itself, and can acquire and manage the start address of the last updated cylinder. The present invention can be implemented even if the update information transfer process is omitted.

[0029]

According to the present invention, even when the host cannot specify the last updated cylinder by using the extended channel command system such as the CKD-EX command system, the disk controller can surely specify the last updated cylinder. Is detected and stored as update information, it is possible to recover a duplicated volume with a minimum amount of data copying when a failure occurs.

[Brief description of the drawings]

FIG. 1 is a diagram showing the basic configuration of the present invention.

The [2] duplexing volume processing in the present invention embodiment
It is a block diagram of the input-output device which performs .

FIG. 3 is a configuration diagram of an embodiment of a disk control device of the present invention.

FIG. 4 is an explanatory diagram of an example of update information in the embodiment of the present invention.

FIG. 5 shows an input / output device that performs conventional duplex volume processing.
It is a block diagram of.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-1-118921 (JP, A) JP-A-3-57047 (JP, A) JP-A-59-168558 (JP, A) JP-A-59-168 176856 (JP, A) JP-A-61-253676 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 3/06

Claims (1)

(57) [Claims] 1. A plurality of devices each including a volume logically treated as one input / output device, a host and a plurality of devices connected to each other, and performing data transfer between each of the volumes and the host. An I / O device comprising: a disk controller ; and the disk controller updates update information corresponding to each volume with update start position information and update information.
Cylinder switching occurs while writing data to a non-volatile storage medium and volume using the end position information
A cylinder switching detection mechanism that detects the occurrence of an update, and an update start address when a write command is executed from the host.
Based on the scan update the updated start position information of the update information corresponding to the volume stored in the nonvolatile storage medium, the cylinder switch detecting mechanism during data write
Each time cylinder switching is detected, the update information
Output apparatus characterized by comprising: means Ru rewriting the update end position information.