JPH11249820A - Disk control system and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten a processing time and to reduce a re-transmission processing time when a fault occurs to min. by continuously transmitting plural tracks, recognizing the responses when data is backed-up to a sub-center from the side of the main center with a line and reproducing only data of the track whose response is not recognized when the fault occurs. SOLUTION: FCU at the side of the main center 1 corresponds to the data updating requests of the plural tracks from HOST, divides the plural tracks into packets and continuously transmits them to the sub-center 2 with the line. FCU at the side of the sub-center 2 repeats receiving the transmitted packets, assembling them into the original tracks and transmitting the responses to the main center 1 whenever assembling is completed. In this case, the track whose response is not received even after a prescribed time elapses among the plural tracks which are transmitted at the side of the main center 1 is detected, the detected track is divided into the packets and they are continuously transmitted to the sub-center 2 again with the line.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a disk control system and a recording medium for transferring data of a primary center to a secondary center via a line.

[0002]

2. Description of the Related Art Conventionally, when a computer center of a company (referred to as a main center) is damaged by an earthquake or the like, another computer center (referred to as a sub-center) provided in a remote place is required to perform computer operations. Magnetic disk unit (DASD) at the center
The file to be written to the sub-center DAS via the line
D is backing up. Thus, when a file required to be backed up to the sub-center is updated from the host computer of the main center (referred to as host) to the DASD by the magnetic disk controller (referred to as FCU) of the main center, the update is performed via the line. To the FCU of the sub-center to send the DAS of the sub-center.
The same update is performed for D, and the DASD of the main center is always
And the contents of the DASD of the sub-center are matched.

[0003]

At this time, conventionally, FIG.
From the main center to the sub-center by the procedure shown in 3
The transmission and response of data to be written to the ASD have been performed. Hereinafter, the configuration and operation will be described in detail.

FIG. 13 shows an explanatory diagram of the prior art. FIG.
In 3, the main center and the sub-center are configured by a CH and an FCU, as shown in the figure.

[0005] The CH is a channel device for transmitting and receiving data between the host and the FCU. FC
U is a magnetic disk control device for controlling access to the magnetic disk device and transmitting / receiving data to / from a partner center via a line.
It is composed of RA and the like.

[0006] CA is a channel adapter, CH
Write the data received from the SS to the SS, or pass the data of the SS to the CH. The SS is a non-volatile memory that temporarily holds data.

[0007] RA is a remote adapter for transmitting and receiving data to and from a partner RA via a line. Next, a procedure for backing up data from the primary center to the secondary center will be described with reference to FIG.

(1) The host on the main center side is CCW D
The X, CCW LR notifies the FCU of the update process, the location information of the updated track, the number of tracks to be updated, and the like.
Here, it is assumed that three tracks (TRK0, 1, 2) have been updated.

(2) According to CCW WUD, CA becomes C
Update data is received from H and written to SS. (3) After storing in the SS for each track, the update data is divided into packets (for example, one track is divided into four packets) by the RA and transmitted to the RA of the sub-center.

(4) The RA of the sub center receives the packet transmitted in (3) and assembles it as the original track.
Write to. In this case, one track is divided into four
There is a line delay time from the start of transmission of the packets P0 to P3 to the RA of the sub-center continuously until the RA of the sub-center starts receiving packets. When measured on an actual device, there is a line delay time of about 10 ms.

(5) When the RA of the sub-center completes receiving all packets (4 in this case) in (4), it transmits a response packet to the main center. (6) The main center receives the response packet transmitted in (5) and recognizes that the transfer processing of the track 0 has been completed normally. Similarly, a line delay time (approximately 10 ms on an actual device) is generated until the response packet is received.

(7) Hereinafter, the same is repeated for tracks 1 and 2. The above procedure is shown in FIG. 13, and the data of three tracks (TRK0, 1, 2) is transferred from the main center to the subcenter via the line and the response is confirmed. In addition, line delay time for 6 times, about 10m
There is a problem that a processing time delay of s × 6 = 60 ms occurs.

In particular, as described above, if the line delay time caused by the three-track transfer is about 60 ms due to the delay, about 40% of the total processing time for transferring data from the main center to the sub-center via the line is occupied. As a result, there is a problem that the processing efficiency is extremely poor.

[0014] The present invention solves these problems.
When data is backed up from the main center to the sub-center via the line, the number of tracks is continuously transmitted and the response is checked, and only the data of the track whose response cannot be confirmed when a failure occurs is retransmitted. It is an object of the present invention to reduce the processing time by improving the processing efficiency and to minimize the retransmission processing time when a failure occurs.

[0015]

Means for solving the problem will be described with reference to FIG. In FIG. 1, a main center 1 is a center that performs various business processes.
This is a center where data is evacuated to the sub-center 2 and prepared for an earthquake or the like. As shown in the figure, HOST (host computer), FCU (magnetic disk controller), LR (connection device), and DASD (magnetic disk) Storage device).

The sub-center 2 is a center installed at a remote location from the main center 1 that performs various business processes. Here, the data transferred from the main center 1 is stored in the same manner as the main center 1. And, as shown, HO
It comprises an ST (host computer), an FCU (magnetic disk control device), an LR (connection device), and a DASD (magnetic disk storage device).

Next, the operation will be described. F on the center 1 side
The CU divides a plurality of tracks into packets in response to a request for updating data of a plurality of tracks from the HOST and continuously transmits the packets to the sub-center 2 via a line.
U receives the transmitted packet, assembles it into the original track, and repeatedly sends a response to the main center 1 each time the assembly is completed, and the FCU on the main center 1 receives responses for all of the multiple tracks. When it is done, it is confirmed that transmission is completed.

At this time, a track for which no response is received even after a predetermined time has elapsed among a plurality of tracks transmitted by the main center 1 is detected, and the detected track is divided into packets and continuously transmitted via a line. The data is transmitted to the sub center 1 again.

Further, in response to the request for updating the data of a plurality of tracks from the HOST, the FCU on the main center 1 divides the plurality of tracks into packets and continuously transmits them to the sub-center 2 via a line. The FCU of the second side receives the transmitted packet and assembles it into the original track, and sends a response to the main center 1 when all the trucks have been assembled,
When the response is received by the main center 1, it is confirmed that the transmission is completed.

At this time, when a response is received by the main center 1, a non-responding track is detected based on the response information, the detected track is divided into packets, and the detected track is successively transmitted via the line. The data is transmitted to the center 2 again.

[0021] Also, of the plurality of tracks transmitted on the main center 1 side, a track for which a response is not received even after a lapse of a predetermined time is detected, or when the main center 1 side receives a response, the response information is also transmitted. Then, the unresponsive track is detected, the detected track is divided into packets, and the packets are continuously transmitted to the sub-center again via another line.

A CCW (channel command word) for setting a switching flag indicating that a plurality of tracks are collected from the primary center 1 to the secondary center 2 or divided into packets for each track and transmitted continuously via a line. ) Is provided so that the FCU of the primary center 1 collectively divides a plurality of tracks according to the switching flag of the CCW or divides it into packets for each track and continuously transmits the packets to the secondary center 2 via a line.

Therefore, when data is backed up from the main center 1 to the sub-center 2 via the line, the data is continuously transmitted for a plurality of tracks and the response is confirmed. By performing retransmission, it is possible to improve the reduction in processing efficiency due to the line delay time, to shorten the processing time, and to minimize the retransmission processing time when a failure occurs.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments and operations of the present invention will be sequentially described in detail with reference to FIGS.

FIG. 1 shows a system configuration diagram of the present invention.
FIG. 1A shows an overall system configuration diagram. In FIG. 1A, the main center 1 includes HOST, FCU, L
R and DASD.

The HOST is a host computer for performing various business processes. The FCU is a magnetic disk controller that controls access to the DASD and transfers data to be evacuated to the sub-center 2 in response to an access request by the CCW from the HOST. It is configured as shown in b).

The LR is a connection device for transmitting and receiving data to and from the LR of the sub center 2 via a line. DASD is a magnetic disk storage device that stores data in tracks.

The communication line is a line for transmitting and receiving data between the main center 1 and the remote center 2.
The sub center 2 receives the data transfer from the main center 1 and stores the same contents as the main center 1 in DASD,
In the event of a failure, the business process is executed on behalf of
It has the same configuration as the main center 1, LR, FC
U, HOST, and DASD.

FIG. 1B shows a detailed system configuration diagram of the FCU of FIG. In FIG. 1B, CA is a channel adapter and is a functional unit connected to a higher-level device (HOST).

RM is a resource manager, F
It is a functional unit that controls the CU internal resources. The CFE is a cache function engine, and is a functional unit for performing cache control.

SS is a cache memory, which is a nonvolatile memory for temporarily storing data. TS
Is a table storage, which is a memory for forming a control table.

RA is a remote adapter, and is a functional unit connected to a line. DA is a device adapter, and is a functional unit connected to a magnetic disk storage device.

C-BUS is an internal bus. (C) of FIG. 1 schematically shows a data buffer (Buffer 0) inside a DRA (functional unit connected to a line).
In the present invention, tracks 0 to n are stored in the DRA.
In contrast, only one Buffer 0 is provided, data of a plurality of tracks is stored in this buffer 0, and output to a plurality of tracks at the time of output.

Next, the operation of the configuration of FIG. 1 will be described in detail according to the order of the flowchart of FIG. FIG. 2 is a flowchart illustrating the operation of the present invention.

In FIG. 2, in S1, a file update request is generated. In S2, it is determined whether or not a plurality of tracks are updated. Y
In the case of ES, the process proceeds to S3 and subsequent steps according to the present invention. In the case of NO, it is a one-track update, which is not the processing targeted by the present invention, and thus ends (the conventional one-track update processing is performed).

At S3, the HOST issues a CCW DX. At S4, the FCU receives the CCW DX. S5
HOST issues CCW LR.

At S6, the FCU receives the CCW LR. Since these S3, S4, S5, and S6 are a multi-track update at YES in S2 and are determined to be the target processing of the present invention, the HOST is, for example, a CC shown in FIG.
Set the backup flag to “ON” in WDX, FIG.
(B) of the operation code to the CCW LR "WRIT
E ", update position is" Track 0 ", update track
Number / Record Number is set to “Number of Records included in Tracks 0 to 2”, respectively, and these are set to FCU.
And confirms the backup (writing to DASD and transferring it to subcenter 2 for backup at the same time) and updating of a plurality of tracks (Tracks 0 to 2).

In S7, the HOST issues a CCW WUD. S8 is stored in SS. This is HOS in S7
C that configures the FCU with CCW WUD issued by T
A receives it and stores it in SS.

At S9, the data is transmitted to the sub center 2. S10
Determines whether the continuous transmission is completed for tracks 0 to 2. Y
In the case of the ES, when the response reception of the tracks 0 to 2 is confirmed in S11, a response is notified to the host in S12. On the other hand, if NO in S10, the process returns to S7 and repeats. By repeating S7 to S10, here, for example, each track of tracks 0 to 2
This makes it possible to divide the packet into a plurality of packets and continuously transmit them to the sub center 2 collectively (total 12 packets for three tracks are transmitted continuously and collectively).

In S13, the sub center 2 sequentially and sequentially receives the packets transmitted from the main center 1 in S9. S
Reference numeral 14 reassembles a packet received by the FCU CA of the sub-center 2 at the SS, stores the packet in a track, and stores the track in the SS.

In step S15, a response is transmitted. Each time truck assembly is completed in S14 and stored in SS, tracks 0 to 2
Is transmitted to the main center 1 (this confirmation response is received and confirmed by the main center 1 as a confirmation response for each track in S11 described above). And S1
Return to 3 and repeat.

As described above, the main center 1 collectively divides a plurality of tracks (here, tracks 0 to 2) into 12 packets and successively transmits the packets to the sub-center 2, and the sub-center 2 sequentially receives the packets. When the acknowledgment is sent to the main center 1 for each of the tracks received for the tracks 0 to 2 and stored in the SS, the transfer for a series of backups is performed when the acknowledgment is received for all the tracks at the main center 1. Recognizes completion, and ends. By dividing a plurality of tracks into packets and transmitting them continuously to the sub-center 2 in this manner, as shown in an explanatory diagram of FIG. The transmission line delay time is the first and last acknowledgment line delay time, and the intervening line delay time is included in the transfer time. As a result, the processing time delay is If the delay time is doubled, that is, if the line delay time is 10 ms measured on a real machine, the delay time is 1
0 ms × 2 = 20 ms, which can be reduced to about ３ as compared with the conventional 10 ms × 6 = 60 ms in the case of FIG.

FIG. 3 shows a CCW example of the present invention. FIG. 3A shows an example of CCW DX. This CCW DX
Is an example of the CCW issued by the HOST to the FCU in S3 of FIG. 2 described above, and is for setting and notifying the following items shown in the figure.

Backup Flag Switching Flag ON / OFF ON / OFF Here, the backup flag is a flag indicating whether or not the update data needs to be backed up from the main center 1 to the sub center 2. The switching flag is shown in FIG.
As used in the flowchart of FIG. 2, when backing up, a mode in which a plurality of tracks according to the present invention are divided into packets and transmitted to the sub center 2 continuously (when the switching flag is ON), To be transmitted to the sub-center 2 (the conventional case in which the switching flag is OFF).

FIG. 3B shows an example of CCW LR.
The CCW LR is an example of the CCW issued by the HOST to the FCU in S5 of FIG. 2 described above, and is set and notified as follows.

Operation code Update position Update record number WRITE / READ Track 0 Record number included in Tracks 0 to 2 Here, the operation mode is WRITE / REA for DASD.
D. The update position is a position (Track) where the update is started at the time of WRITE. Update Record
The number (the number of updated Tracks) is the Record to be updated.
(Track) number.

The above CCW DX and CCW LR are replaced by HO
The FCU receives the ST from the ST. The FCU determines whether or not to backup to the sub-center 2 based on the backup flag, whether or not to transmit a plurality of tracks collectively or conventionally divided into packets for each track according to the switching flag, and the update position. ,
It is possible to recognize the number of update records / tracks.

FIG. 4 is an explanatory diagram (response for each track) of the present invention. This is a multitrack (TRK in this example)
0 to 2) are respectively divided into four packets P0 to P4 and transmitted continuously, and the sub-center 2 on the receiving side sequentially transmits acknowledgments PR0 to PR2 to the main center 1 every time the truck is assembled. This is an example of the case. Here, the symbols represent the following, respectively.

DX is the CCW DX of S3 in FIG.
(A)). LR is the CCW LR of S5 in FIG.
((B) of FIG. 3). WUD is the CC of S7 in FIG.
Represents W WUD.

TRK0-TRK2 represent tracks 0-2. P0 to P3 represent packets obtained by dividing TRK0 to TRK2 into four, respectively.

PR0 to PR2 indicate responses in which confirmation of reception of TRK0 to TRK2 by the sub center 2 has been made. The line delay time is a delay time from when the packet is transmitted from the main center 1 to reach the sub center 2, and the line delay time is a delay time from when the packet reaches the sub center 2 to the main center 1. Actually measured at about 10
This is a delay time on the order of ms. Here, the line delay time is included during transmission of a packet, and does not cause a delay in the entire processing time, but delays the entire processing time only in the line delay time and the line delay time. A delay time of 10 ms × 2 = 20 ms occurs (a delay time of 40 ms is shorter than that of 10 ms × 6 = 60 ms in the conventional case of FIG. 4).

As shown in FIG. 4, TRK0 to TRK2 are divided into packets P0 to P3, respectively, and transmitted from the primary center 1 to the secondary center 2.
3 each time one track is assembled
By sequentially replying to PR2, the primary center 1 can recognize that the data to the secondary center 2 has been completed, and
At this time, the total delay time can be reduced to 20 ms, that is, 60-20 = 40 ms as compared with the conventional case of FIG.

FIG. 5 is a flowchart illustrating another operation of the present invention. 5, S21 corresponds to B (S1 in FIG. 2).
In the case of YES of 0 (in the case where the data is divided into packets and continuously transmitted to the sub-centers 2 from the tracks 0 to 2), the response from the sub-center 2 is received (all received). This is S2
In step 6, all tracks from the sub center (here, tracks 0 to
In response to receiving 2), the main center 1 receives a response.

In step S22, a response is sent to the host of the main center 1. In S23, in the case of A in FIG. 2 (when the main center 1 continuously transmits the packets obtained by dividing the tracks 0 to 2 to the sub-center 2 in S9 in FIG. 2), the sub-center 2 receives the packet.

In step S24, the data is stored in the SS. In step S25, it is determined whether the process has been completed up to the track 2. If yes,
Since all the tracks have been received, a response is transmitted to the main center 1 in S26, and the process ends. On the other hand, in the case of NO, S2
Return to 3 and repeat.

As described above, the main center 1 collectively divides a plurality of tracks (here, tracks 0 to 2) into twelve packets and successively transmits them to the sub center 2, and the sub center 2 receives the packets sequentially. When assembled in a truck and stored in the SS, an acknowledgment was transmitted to the main center 1 when all of the tracks 0 to 2 were received, and when the acknowledgment was received by the main center 1, the transfer was completed for a series of backups And notifies the host and terminates. As described above, by dividing a plurality of tracks into packets and continuously transmitting them to the sub-center 2, as shown in an explanatory diagram of FIG. Main center 1 delay time and PR (acknowledge)
When the line delay time is twice as long as the line delay time, that is, when the line delay time is 10 ms measured on a real machine, the delay time is 10 ms × 2 = 20 ms, which is a conventional delay time. FIG. 13 already described
In this case, it is possible to reduce the time to about 1/3 of 10 ms × 6 = 60 ms.

FIG. 6 is an explanatory diagram (response of all tracks collectively) according to the present invention. This is done by dividing all the tracks (TRK0 to TRK2 in this example) into four packets P0 to P4 and transmitting them continuously, and confirms when the sub-center 2 on the receiving side has assembled all the tracks. This is an example of a case where a response PR is transmitted to the main center 1. Here, the symbols represent the following, respectively.

DX is the CCW DX of S3 in FIG. 2 (FIG. 3).
(A)). LR is the CCW LR of S5 in FIG.
((B) of FIG. 3). WUD is the CC of S7 in FIG.
Represents W WUD.

TRK0-TRK2 represent tracks 0-2. P0 to P3 represent packets obtained by dividing TRK0 to TRK2 into four, respectively.

The PR sets TRK0 to TRK2 to the sub-center 2
Indicates a response for which all the data has been confirmed. The line delay time is a delay time from when a packet is transmitted from the main center 1 to reach the sub-center 2, and the line delay time is a delay time from when a response arrives from the sub-center 2 to the main center 1. About 1 when actually measured on actual machine
This is a delay time of about 0 ms. Here, there are two delays of the entire processing time, the line delay time and the line delay time, and a delay time of 10 ms × 2 = 20 ms occurs (10 ms × 6 = the conventional case of FIG. 4 in FIG. 4).
The delay time is reduced by 40 ms compared to 60 ms).

As shown in FIG. 6, TRK0 to TRK2 are divided into packets P0 to P3, respectively, and transmitted from the primary center 1 to the secondary center 2.
13 by assembling all the tracks and returning a response PR, the primary center 1 can recognize that the data to the secondary center 2 has been completed, and the total delay time at this time is 20 ms, which is the case of FIG. 60-2 compared to
0 = 40 ms can be shortened.

FIG. 7 is a flowchart illustrating another operation of the present invention (retransmission when a failure occurs, part 1). S31 is
Failure occurs. This is because the acknowledgment transmitted by the sub center 2 in S15 of FIG. 2 described above cannot be received by the main center 1 in S11, and some trouble occurs on the line.

In S32, a timeout occurs. This is because the main center 1 cannot receive the acknowledgment of any of the tracks in S11 of FIG. S33
Extracts untransmitted tracks.

At S34, retransmission is performed. This is because a track which has timed out in S32 and for which an acknowledgment has not been received is extracted, and the track is divided into packets from the main center 1 to the sub-center 2 only for the relevant track and transmitted continuously,
The main center 1 receives the confirmation response and confirms that the transmission has been completed.

As described above, according to the flowchart of FIG. 2, a plurality of tracks are divided into packets and transmitted from the primary center 1 to the secondary center 2, and the secondary center 2 sends an acknowledgment for each track within the specified time. When a time-out occurs, only the unacknowledged tracks are retransmitted from the main center 1 to the sub-center 2 and the acknowledgments are received and a series of backup processing is completed. Only the efficient retransmission can guarantee the integrity of the backup processing.

FIG. 8 is an explanatory diagram of the present invention (retransmission when a failure occurs, part 1). According to this, according to the flowchart of FIG. 7 described above, the track is divided into a plurality of packets P0 to P3 and transmitted continuously from the primary center 1 to the secondary center 2,
An acknowledgment PR0 every time a truck can be assembled at the sub center 2.
To PR2 are sequentially transmitted to the main center 1. The main center 1 can receive only PR0, cannot receive PR1 and PR2, and time-out occurs. TRK1 and TRK2 corresponding to PR1 and PR2 are retransmitted to correct PR1 and PR2. This shows a procedure when the center 1 receives and confirms the data and completes a series of retransmission processing.

FIG. 9 is a flow chart for explaining another operation of the present invention (retransmission when a failure occurs, part 2). S41 is
Failure occurs.

In step S42, a response is received. A step S43 decides whether or not there is a missing response. This is because the TRKs 0 to 2 are divided into packets and transmitted continuously from the main center 1 to the sub center 2, and the sub center 2 assembles all the tracks (TR
When K0-2) is received, the acknowledgment PR is sent to the main center 1
Therefore, it is determined whether there is a missing track among the tracks transmitted during the acknowledgment in the main center 1.

A step S44 extracts untransmitted tracks. S
45 resends. As described above, in accordance with the flowchart of FIG. 5, a plurality of tracks are divided into packets and transmitted from the primary center 1 to the secondary center 2. When the secondary center 2 receives acknowledgments of all the tracks, the primary center 1 By extracting missing tracks, retransmitting them, receiving an acknowledgment, and ending a series of backup processing, it is possible to efficiently retransmit only untransmitted tracks and to guarantee the integrity of the backup processing. .

FIG. 14 shows an explanatory diagram of the present invention. This shows a situation where the main center recognizes, for example, PR1 which has disappeared at the time of occurrence of the failure in FIG. 9 and retransmits the packet. FIG.
0 is another operation explanation flowchart of the present invention (HOT /
STBY retransmission). As shown in the configuration diagram of FIG. 11A, which is described later, the primary center 1 uses the RA0 as the HOT (working system) and retransmits it to the subcenter 2, and when a failure occurs, the RA1 is used as the STBY (standby system). 5 is a flowchart when retransmission is performed from the main center 1 to the sub center 2 by using. In this case, the HOT notifies STBY of “STATUS” (information of a track for which a response cannot be confirmed at the time of failure occurrence) to the STBY.
It is characterized in that only tracks whose response cannot be confirmed based on TUS are retransmitted.

In FIG. 10, in S51, retransmission is performed by HOT (working system). This corresponds to H of FIG.
The track is divided into a plurality of packets using RA0 of the OT (working system) and retransmitted from the primary center 1 to the secondary center 2.

A step S52 decides whether or not the number is exceeded. This is done by dividing the track into a plurality of packets by using the RA0 of the HOT in S51, and determining whether the number of times that the time has been retransmitted from the primary center 1 to the secondary center 2 exceeds the specified number. In the case of YES, the process proceeds to S53. If NO, S51 and S52 are repeated.

In S53, the retransmission is not possible in the HOT by the YES in S52, so that retransmission is performed by STBY (standby system). In this method, a track is divided into a plurality of packets by using the STBY RA1, and the packet is retransmitted from the primary center 1 to the secondary center 2.

A step S54 decides whether or not the number of times is exceeded. YE
In the case of S, the STBY system also ends as a failure occurrence.
In the case of NO, S53 and S54 are repeated. As described above, retransmission is performed from the main center 1 to the sub center 2 using RA0 of the HOT (working system).
By performing retransmission from the primary center 1 to the secondary center 2 using RA1 of the BY (standby system), RA0 and RA1,
Furthermore, it is possible to perform retransmission by changing the line, improve the success rate of retransmission due to the occurrence of a line failure, and improve reliability.

FIG. 11 is an explanatory diagram (HOT / ST) of the present invention.
BY retransmission). FIG. 11A shows a configuration diagram. The configuration shown in the figure is equivalent to the FCU shown in FIG.
RA1 is added, and RA0 is used in the HOT (active system) and RA1 is used in the STBY (standby system), so-called duplication is achieved.

FIG. 11 (b) is a diagram showing the case where H0 RA0 to SOT
Switch to TBY RA1 and switch from primary center 1 to secondary center 2.
Shows the situation when resending to. The upper part divides the tracks TR0 to TR2 into packets using the RAOT of the HOT of the main center 1 and continuously transmits the packets to the subcenter 2, and transmits the responses PR0 to PR2 to the main center 1 for each track. It shows how to do. At this time, PR1 and PR2 have been lost due to a rotation failure or the like.
Tracks TR1 and TR2 are divided into packets using RA1 of TBY, and are continuously retransmitted to sub-center 2.
1 and 2 are transmitted to the main center 1 on a track-by-track basis (in the figure, the process up to transmission of P0 to P3 of TRK1 is shown, and the rest is omitted).

As described above, the RA of the HOT of the primary center 1
0, and when retransmission is not possible due to a line failure or the like, retransmission is performed using RA1 of STBY, so that HO
It is possible to enhance reliability by so-called duplication of T / STBY.

FIG. 12 is a flowchart (switching flag) for explaining another operation of the present invention. This explains the operation when the switching flag is turned ON / OFF in the CCW DX of FIG. 3A described above.

In FIG. 12, in S61, a file update occurs. A step S62 decides whether or not the switching flag is ON.
That is, the FCU that has received the CCW DX from the host determines whether the switching flag in the CCW DX is ON. YE
In the case of S, S2 and subsequent steps of F of FIG.
A plurality of tracks are divided into packets and transmitted continuously, and the received sub-center 2 transmits an acknowledgment to the main center 1 to recognize completion of a series of backup processing. On the other hand, in the case of NO, the response is confirmed for each conventional track in S63 (see FIG. 13 and the description of the conventional technique described above).

As described above, the host can execute the processing according to the present invention after S2 in FIG. 2 described above, or can execute the processing in the related art described above only by setting the switching flag in the CCW DX to ON / OFF. For example, it is possible to execute a process of transmitting each track shown in FIG.

[0081]

As described above, according to the present invention,
When backing up data from the main center 1 to the sub-center 2 via a line, a configuration is adopted in which multiple tracks are continuously transmitted and the response is confirmed, and only data on the track for which a response cannot be confirmed when a failure occurs is retransmitted. As a result, it is possible to reduce the processing time by reducing the processing efficiency due to the line delay time, and to minimize the retransmission processing time when a failure occurs.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of the present invention.

FIG. 2 is a flowchart illustrating the operation of the present invention.

FIG. 3 is a CCW example of the present invention.

FIG. 4 is an explanatory diagram (response for each track) of the present invention.

FIG. 5 is a flowchart illustrating another operation of the present invention.

FIG. 6 is an explanatory diagram of the present invention (responses collectively for a plurality of tracks).

FIG. 7 is a flowchart illustrating another operation of the present invention (retransmission when a failure occurs, part 1).

FIG. 8 is an explanatory diagram of the present invention (retransmission when a failure occurs, part 1)
It is.

FIG. 9 is a flowchart illustrating another operation of the present invention (retransmission when a failure occurs, part 2).

FIG. 10 is a flowchart (HO) for explaining another operation of the present invention.
T / STBY retransmission).

FIG. 11 is an explanatory diagram of the present invention (retransmission of HOT / STBY).
It is.

FIG. 12 is a flowchart (switching flag) for explaining another operation of the present invention.

FIG. 13 is an explanatory diagram of a conventional technique.

FIG. 14 is an explanatory diagram of the present invention.

[Explanation of symbols]

 1: Primary center 2: Secondary center HOST: Host (host computer) FCU: Magnetic disk controller LR: Connection device DASD: Magnetic disk device CA: Channel adapter RM: Resource manager SS: Cash memory RA: Remote adapter RA0: HOT ( Active RA) RA1: STBY (Standby) RA DA: Device Adapter

Claims (8)

[Claims] 1. A disk control system for transferring data of a main center to a sub-center via a line, means for dividing a plurality of tracks of the main center into packets, and successively transmitting the packets to the sub-center via a line, Means for receiving the transmitted packet at the sub-center, assembling it into the original track, and transmitting a response to the main center each time the assembling is completed; A disk control system comprising: means for confirming transmission completion when received. 2. The main center detects a track of which the response has not been received within a predetermined period of time among the plurality of transmitted tracks, and divides the detected track into packets to continuously connect the line. 2. The disk control system according to claim 1, further comprising means for transmitting data to the sub-center again via the sub-center. 3. A disk control system for transferring data of a main center to a sub-center via a line, means for dividing a plurality of tracks of the main center into packets and transmitting the packets to the sub-center via the line continuously, Means for receiving the transmitted packet at the sub-center and assembling it into the original track, transmitting a response to the main center when all the trucks have been assembled, and transmitting when the response is received at the main center A disk control system comprising: means for confirming completion. 4. When the response is received by the main center, an unresponsive track is detected based on the response information,
4. The disk control system according to claim 1, further comprising means for dividing the detected track into packets and continuously transmitting the divided packets to the sub-center again via a line. . 5. The main center side detects a track of which the response is not received within a predetermined time out of the plurality of transmitted tracks, or the main center side receives the response information when the main center side receives the response. A means for detecting a non-responding track based on the received data, dividing the detected track into packets, and continuously transmitting the divided packets to the sub-center again via another line. The disk control system according to claim 1 or 3. 6. A CCW (channel command word) for setting a switching flag indicating that the plurality of tracks are collectively transferred from the main center to the sub-center or divided into packets for each track and transmitted continuously via a line. 6. The method according to claim 1, wherein a plurality of tracks are combined according to the CCW switching flag or divided into packets for each track, and transmitted to the sub-center via a line continuously. A disk control system according to 7. A means for dividing a plurality of tracks of the main center into packets and transmitting the packets to the sub-center continuously via a line, receiving the transmitted packets at the sub-center and assembling them into the original track, A computer that records a program for functioning as a means for transmitting a response to the main center every time assembly is completed, and a means for confirming that transmission has been completed when the response is received for all of the plurality of tracks transmitted on the main center side A readable recording medium. 8. A means for dividing a plurality of tracks at the main center into packets and transmitting the packets to the sub-center continuously via a line, receiving the transmitted packets at the sub-center and assembling them into the original track, A computer-readable recording medium storing a program functioning as a means for transmitting a response to the main center when the truck assembly is completed and a means for confirming the completion of transmission when the main center receives the response.