JP2003050675A - Data multiplexing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize data multiplexing by to remote copying without causing the deterioration of performance of normal host I/O processing or the like. SOLUTION: This data multiplexing system is constituted of a main system constituted of an external storage controller 3 connected to a channel device 2 and an external storage device 7 and a sub-system constituted of an external storage controller 10 connected to a channel device 9 and an external storage device 12. Also, this information processing system is provided with a data storage device for receiving update data in the other information processing system through an information communication medium. This data multiplexing system is provided with an operation information table 21A for recording a final time TL when the update data are received.

Description

Description: BACKGROUND OF THE INVENTION [0001] 1. Field of the Invention [0002] The present invention relates to a data multiplexing technique, and more particularly, to an arbitrary information communication medium between a plurality of information processing systems installed at remote locations. The present invention relates to a technology that is effective when applied to a technology for improving the reliability and maintainability of data by remotely copying data, multiplexing the data, and holding the data. 2. Description of the Related Art As a technique for improving data reliability and the like, for example, a RAID technique is known. However, in order to reliably avoid data loss due to a disaster such as a fire or an earthquake. However, multiplexing data within the same site is not sufficient. For example, it is essential to multiplex data between a plurality of data storage devices installed at remote locations. As a data multiplexing method for holding data from a host device in a plurality of external storage devices, an external storage device holding the same data is provided in each of the primary and secondary systems, and the external storage device of each of the primary and secondary systems is provided. It is conceivable to connect the control devices with an information communication medium or the like. In such a system, a command similar to the command received from the primary external storage control device that has received the write command from the higher-order device to the secondary external storage control device connected thereto is transmitted to the primary storage control device. A method of matching data in the primary and secondary systems by issuing the command in synchronization with command reception is conceivable. Alternatively, it is conceivable to match data by issuing a command to the sub system side asynchronously without synchronizing with the command received by the main system side. [0004] As a prior art relating to such data multiplexing, for example, EMC, published in November 1995,
"Product Description Guide
e) and the like, when a command from a higher-level device is received by the primary external storage controller, only the channel end is reported to the higher-level device, and data is transferred to the secondary external storage controller to perform data transfer. A technique for matching the data in the external storage devices under the primary and secondary systems by reporting the device end to the host device after the transfer is completed, and when receiving a command from the host device, reporting the channel end / device end to the host device, There is known a technique for performing data transfer between primary and secondary external storage controllers to achieve data matching between external storage devices under the primary and secondary systems. In the conventional system as described above, the same data as the write data received by the primary external storage control device is transferred to the external storage device under the secondary external storage control device. Considering the process of writing to the external storage device, the completion report to the channel cannot be made until after the writing to the external storage device under the secondary external storage control device is completed, so that the response time to the channel increases. Also, since the primary and secondary external storage controllers are connected by an interface cable, there are few paths to channels that can be used for data transfer (host I / O) with a higher-level device. If a write command is always issued to the secondary side in response to the reception of the write command, there is a concern that the throughput of the host I / O processing is reduced because the system resources (processor, cache memory, etc.) are used by the data transfer processing. Further, since the secondary system is a backup system, it is supposed that the external storage controllers are disposed at a long distance by using an optical fiber cable in case of a disaster in case of a disaster. And the effect on the host I / O increases accordingly. As a countermeasure for this problem, it is conceivable that data is asynchronously written to the secondary external storage device without being synchronized with the write command from the host device. According to this method, the data transfer to the secondary side is performed by the original I
Since it can be executed in the background of the / O process, the response time to the write command is faster than in the case where a command for data copying is issued to the secondary side in synchronization with the write command from the higher-level device. However, in the method of executing data transfer asynchronously, depending on the timing of data transfer, data transfer may be started when the primary system is under heavy load. There is a concern that overall performance will be degraded due to the use of resources. Also, in the asynchronous transfer, there is a time when the data of the primary and secondary systems do not match, so there is a problem in data assurance in the event of a failure during that time. An object of the present invention is to provide a data multiplexing technique capable of realizing data multiplexing by remote copy while minimizing a decrease in the information processing capability inherent in the system. Another object of the present invention is to realize data multiplexing by remote copy using a plurality of data storage devices installed at remote locations while minimizing a decrease in the information processing capability inherent in the system. It is an object of the present invention to provide a data multiplexing technique capable of performing the above. Another object of the present invention is to provide a data multiplexing technique capable of realizing data multiplexing by using a minimum amount of information processing resources by taking into account the magnitude of system load and operating characteristics. To provide. Another object of the present invention is to provide a data multiplexing technique capable of realizing data multiplexing by remote copy at various timings according to a user instruction or the like. Another object of the present invention is to provide a data multiplexing technique capable of realizing highly reliable data multiplexing while minimizing deterioration of the information processing capability inherent in the system. . [0014] The present invention provides a data multiplexing system for realizing data multiplexing by transferring / copying update data between at least two first and second information processing systems. The trigger for starting the copy of the update data is based on the amount of accumulated difference data that is unreflected update data, and the threshold value for determining the amount of the difference data is determined by a user's instruction or the operating status of the system. By variably setting according to the above, flexible, diverse and highly reliable data multiplexing is realized. More specifically, in the case where each of the primary and secondary information processing systems has an external storage device and an external storage control device for storing data exchanged with a higher-level device, for example, Means for variably setting a threshold value of a difference data amount managed on the primary side based on a command received by the external storage control device on the side and information reflecting an operation status such as a usage amount of the primary side cache memory; Means for controlling, by designating an arbitrary data set and a storage device via a higher-level device or a service processor, to selectively transfer data of an update to a specified data range to a secondary side, Allows devices and users to set thresholds for specific operating hours and the amount of difference data used in those operating hours, and sets thresholds for each operating hour. Means for controlling the transfer timing of update data to the secondary side; and transferring differential data not reflected on the secondary side managed by the threshold to the secondary side, in units of data sets, devices, and external storage devices. And means for collectively transferring data for each storage area unit such as a track, and means for recording the last writing time to the secondary side on the secondary side. Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing an example of an information processing system constituting a data multiplexing system according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of each information processing system shown in FIG. FIGS. 3, 4, and 5 are conceptual diagrams showing examples of various control information used in the information processing system of the present embodiment, and FIG. FIG. 7 is a conceptual diagram showing an example of a method for setting a difference data amount threshold in the information processing system according to the embodiment, and FIG. 7 is a flowchart showing an example of the operation of the information processing system according to the present embodiment. As shown in FIG. 1, the primary system, which is an information processing system, includes a central processing unit 1, a channel unit 2, a channel processing execution processor 4, a drive processing execution processor 5, and a cache memory 6. External storage control device 3 and external storage device 7 connected thereto
Consists of The secondary system, which is a backup system of the primary system, comprises a central processing unit 8, a channel device 9, an external storage control device 10, a cache memory 11, and an external storage device 12, like the primary system. The storage controller 3 and the external storage controller 10 of the secondary system are connected by an interface cable 13. When the primary system and the secondary system are arranged at remote locations, the interface cable 13 may be an optical fiber or a communication line connecting them, or an information communication medium such as a radio communication radio wave. Can be In the secondary system on the backup side, the central processing unit 8 and the channel device 9 are not always necessary, but the external storage control device 10 and the external storage device 12
The configuration may be only a combination of the above. Further, as shown in FIG. 2, the external storage control device 3 is provided with an operation information table 21 for managing the operation state of the system and a difference information table 27 for managing the difference data, in addition to the cache memory 6. I have. Further, in the channel processing execution processor 4, a channel command control unit 22 for realizing exchange of processing with the channel device 2, a secondary disk command issuing unit 23 for issuing a command to the secondary system, and difference information for managing primary and secondary difference information A management unit 24 is provided, and the drive processing execution processor 5
Is provided with an external storage device control unit 25 for reading / writing data from / to the external storage device 7. Also,
A service processor 26 (SVP) is connected to the external storage control device 3 of the primary system so that the user can perform operations such as setting a threshold value for differential data. FIG. 3 shows an example of the difference information table 27 provided in the external storage control device 3 of the primary system for managing the difference data. The information stored in the difference information table 27 is information common to the entire system, and gives priority to data transfer (host I / O) between the central processing unit 1 and the channel device 2 by user designation. A set time zone TZ for setting a time zone for suppressing the frequency of data transfer to the secondary system, and a difference data amount which is a threshold value of differential data not reflected in the secondary system managed by the primary system according to user designation. The threshold A, the difference data amount threshold B of the entire system independently determined by the external storage control device 3 when there is no user designation, and the sum of the difference data amounts held in the entire primary system are stored. There is a difference management data total amount Q. For example, when the user instructs a difference data amount threshold value for each data set from the SVP 26 as information for each device, the head address (for example, cylinder) indicating the storage position of the data set in the external storage device 7 No .: head number CCHH), end address (for example, cylinder number: head number CCHH), difference data amount threshold value C in this data set, and difference management data amount q. The operation information table 2 of the secondary system
1A stores a data update last time TL indicating the last update time of the data written from the primary system. FIG. 4 is a difference bitmap table for managing the difference data. The difference bitmap table has a bitmap for each track as an information storage area provided in a storage medium (not shown) constituting the external storage device 7. Bit is O
This indicates that track N has differential data that has not been reflected to the secondary system. FIG. 5 is a conceptual diagram showing an example of the operation information table 21. In the operation information table 21, for example, cache memory usage QC, normal mode RD / WR count N during online processing, sequential mode RD / WR count M corresponding to batch processing, etc. Is recorded. The information for each device (data set) constituting the external storage device 7 includes a normal mode RD / WR count n, a sequential mode RD corresponding to batch processing, and the like.
/ WR number m is recorded. These pieces of information are referred to when the external storage control device 3 uniquely determines the difference data amount threshold value B, as described later. On the other hand, as means for setting the threshold value of the difference data amount, at the time of system construction or after the external storage devices 7 and 12 are formed as primary / secondary pair volumes, the user sets the difference data amount threshold value from the SVP 26 by the user. When the valid time zone of the threshold is designated, this value is written in the area of the set time zone TZ and the area of the difference data amount threshold A in the difference information table 27 in FIG. If the information such as the difference data amount threshold value is not set by the user, data transfer execution to the secondary system for data copying immediately occurs when update data is generated as it is. At this time, if there are many update requests to the external storage device 7 of the primary system, there is a concern that the performance of normal host I / O to the external storage device 7 other than data transfer for data copying may be reduced. Then, the external storage control device 3 independently sets the difference data amount threshold value, and writes this value to the difference data amount threshold value B of the difference information table 27. When setting the difference data amount threshold value for each data set, the difference data amount threshold value C is set for each device. This manages the difference data of the primary system. When the external storage control device 3 reviews the difference data amount threshold value B at a desired timing or at a fixed period,
From the operation information table 21, the cache memory usage QC
And the number of RD / WR (N, M) for each access mode (sequential / normal, etc.) are read. For example, if requests in the sequential mode are concentrated, “batch processing” is performed. If there is a request, the processing contents are separated as in "online processing". Thus, when "online processing", the threshold value of the difference data amount is set large to give priority to the host I / O, and when "batch processing", the data transfer for matching the data of the primary and secondary side systems ( In order to give priority to the (copying) process, the difference data amount threshold value is set small. The difference data amount threshold determined in this way is written to the difference data amount threshold B in FIG. FIG. 6 is a conceptual diagram showing an example of processing when the external storage control device 3 reviews the difference data amount threshold value B at a constant cycle. In FIG. 6, as an example of the determination criterion, the magnitude of the load is determined based on the cache memory usage (case α, β), and the access mode is “batch processing” or “online processing” (case γ, δ), and the magnitude of the set value of the difference data amount threshold value B is determined by a combination of these cases. Hereinafter, an example of the operation of the present embodiment will be described with reference to the flowchart of FIG. Upon receiving the command from the central processing unit 1, the external storage control device 3 of the primary system (step 10)
1), it is determined whether or not it is a write command (step 10)
2) If it is not a write command, a predetermined process corresponding to the command is performed (step 103), and the process returns to step 101. On the other hand, when a write command is received, the following processing is performed. Now, in a state where the external storage device 7 and the external storage device 12 are in a pair state and hold the same data, the external storage device 7 is transmitted from the central processing unit 1 through the channel device 2.
Is issued to the external storage controller 3, the write data is written only in the cache memory 6, an end report for the update request is made, and then the processing is moved to the difference information management unit 24, and the processing is performed in step 104 and after. Is performed. That is, the difference information management unit 24 turns on the corresponding bit of the difference bitmap (track unit) table in the cache memory 6 (step 10).
4), it is determined whether the update request is within the specified data set (step 105). If it is within the designated data set, the management information of the corresponding device is read from the difference information table 27 for each device in FIG. 3 (step 10).
6). The difference information table 27 for each device has a range of a data set and a difference management data amount q for the data set (this value is information set only when the user instructs from the SVP 26), and updates the update data. Is within the designated data set, the difference data amount threshold C of the device is compared with the difference management data amount q (step 107), and the difference data amount threshold C
<= When the difference management data amount becomes q, data transfer to the secondary system is executed (step 108), and the corresponding bit of the difference bitmap (track unit) table is turned off.
(Step 109), and the process returns to step 101. At this time, the secondary system that has received the transfer of the write data stores the reception time of the write data in the operation information table 21A.
Is recorded each time as the data update final time TL. If the value does not exceed the value determined in step 107, the remote copy operation of transferring data to the secondary system is suppressed (step 113), and the difference management data amount q and the difference management data total amount Q are updated. The requested write data amount is added (step 114), and the process returns to step 101. If it is determined in step 105 that the update data is not in the specified data set,
The set time zone TZ is read out with reference to the system common information of the difference information table 27 of
It is determined whether or not Z is included (step 110). If the difference is within the set time, the difference data amount threshold A and the difference management data total amount Q storing the total amount of difference management data managed by the primary system are read from the difference information table 27. (Step 111), the two are compared (Step 112). As a result of the comparison, when the difference data amount threshold value A becomes equal to the difference management data total amount Q, the write data is transferred to the secondary system (step 108), and the corresponding bit of the difference bit map table is turned OFF ( Step 109), and return to step 101. At this time, the secondary system receiving the transfer of the write data stores the reception time of the write data in the operation information table 21A in the data update final time TL.
Record each time. On the other hand, if the difference does not exceed the determination in step 112, the remote copy operation of transferring data to the secondary system is suppressed (step 113), and the total amount Q of difference management data is changed to the write data of the current update request. The amount is added (step 114), and the process returns to step 101. In step 110, if the current time is not included in the set time zone TZ, the difference data amount threshold B and the difference management data total amount Q determined by the primary system are read from the difference information table 27. (Step 115) and compare (Step 116). If the difference data amount threshold value B << = the difference management data total amount Q, the write data is transferred to the secondary system (Step 108) and the difference bit map table The bit is turned off (step 109), and the process returns to step 101. If the value does not exceed the value determined in step 116, the remote copy operation of transferring the write data to the secondary system is suppressed (step 113), and the write data amount of the current update request is added to the total difference management data amount Q. After the addition (Step 114), the process returns to Step 101. If the differential data for which the remote copy has been suspended in step 113 is kept in the cache memory 6 of the primary system, the performance deteriorates. For example, the differential data is temporarily written to the external storage device 7 and later written. Read when executing data transfer to the secondary system. When executing the transfer of the write data in the transfer pending state to the secondary system, the update data held in the same device is collectively referred to by referring to the difference bitmap in the cache memory 6. Forward. The number of transfers can be reduced by transferring the held differential data collectively in device units. Further, since it is necessary to clarify when the write data is reflected in the external storage device 12 of the secondary system, the secondary system, as described above, receives the latest data at the time of receiving the data transfer. By storing the update time in the operation information table 27A, the primary and secondary data update final time TL can be accurately known. Therefore, for example, up to which point data has been duplicated in the primary and secondary systems in the event of a failure or the like can be accurately grasped, and a failure recovery operation can be performed quickly and accurately. As described above, the update data for the primary system is transferred to the secondary system (backup system).
In the process of realizing data multiplexing by transferring the data to the difference system, a difference data amount threshold value (A, B, C) for managing a difference data amount which is update data not reflected in the secondary system from the primary system. Can be set variably, it is possible to accurately and optimally control the timing of data transfer execution for reflecting the difference data to the secondary system according to the operation status of the system. For example, when the load of the normal host I / O processing in the primary system is large, the threshold value of the differential data amount is increased to reduce the frequency of data transfer to the secondary system for remote copy, thereby reducing the load on the primary system. When is small, the threshold value of the difference data amount is reduced, and accurate processing such as increasing the frequency of data transfer to the secondary system for remote copy can be performed. Therefore, for example, it is possible to minimize the influence of the performance degradation that the normal online business on the primary system having a relatively large host I / O load undergoes by executing the backup process for data multiplexing. Further, since the difference data amount threshold value corresponding to each operating time zone, each data set, each device and the like can be set from the user or the upper central processing unit 1, the importance of data is increased. Also, it is possible to realize optimal load control of remote copy (backup) processing according to the system operation schedule and the like. The secondary system records the last reception time of the update data from the primary system, so that when a failure such as a system failure occurs, the update data between the primary and secondary systems is recorded. Can be accurately grasped, and data recovery processing and the like can be performed quickly and accurately. Although the invention made by the present inventor has been specifically described based on the embodiments, the present invention is not limited to the above embodiments and can be variously modified without departing from the gist thereof. Needless to say, there is. According to the data multiplexing system of the present invention, it is possible to realize the data multiplexing by the remote copy while minimizing the deterioration of the information processing capability inherent in the system. Can be Further, there is an effect that data multiplexing by remote copy using a plurality of data storage devices installed at remote locations can be realized while minimizing deterioration of the information processing capability inherent in the system. can get. In addition, by taking into account the magnitude of the system load, the operating characteristics, and the like, it is possible to achieve an effect that data multiplexing by remote copy can be realized using a minimum of information processing resources. Further, there is an effect that data multiplexing by remote copy at various timings can be realized by a user instruction or the like. In addition, there is an effect that highly reliable data multiplexing by remote copy can be realized while minimizing the deterioration of the information processing capability inherent in the system.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an example of an information processing system constituting a data multiplexing system according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating an example of the configuration of each information processing system illustrated in FIG. 1 in further detail; FIG. 3 is a conceptual diagram showing an example of various control information used in the information processing system according to the embodiment of the present invention. FIG. 4 is a conceptual diagram showing an example of various control information used in the information processing system according to the embodiment of the present invention. FIG. 5 is a conceptual diagram showing an example of various control information used in the information processing system according to one embodiment of the present invention. FIG. 6 is a conceptual diagram showing an example of a method of setting a difference data amount threshold value in an information processing system according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating an example of an operation of the information processing system according to the embodiment of the present invention; DESCRIPTION OF SYMBOLS 1 ... Central processing unit, 2 ... Channel device, 3 ... External storage control device, 4 ... Channel processing execution processor, 5 ... Drive processing execution processor, 6 ... Cache memory, 7 ... External storage device (first) 8: central processing unit, 9: channel device, 10: external storage control device, 11
... Cache memory, 12 ... External storage device (second data storage device), 13 ... Interface cable (information communication medium), 21 ... Operation information table, 21A ... Operation information table, 22 ... Channel command control unit, 23 ... Sub Disk command issuing unit, 24 ... difference information management unit, 25 ...
External storage device control unit 26 Service processor 27
... difference information table (threshold setting means), 27A ... operation information table, A ... difference data amount threshold, B ... difference data amount threshold, C ... difference data amount threshold, M ... sequential mode RD / WR times, N: Normal mode RD / WR times, Q: Difference management data total amount, QC: Cache memory usage amount, TL: Data update last time, T
Z: Set time zone, m: Sequential mode RD / WR
Number of times, n: number of times of normal mode RD / WR, q: amount of difference management data.

   ────────────────────────────────────────────────── ─── Continuation of front page    F term (reference) 5B065 BA01 CA12 CE11 CH01 EA02                       EA35                 5B082 DE06

Claims (1)

Claims 1. An information processing system including a data storage device for receiving update data in another information processing system via an information communication medium, the last time the update data was received. Data multiplexing system to record.