JPH04271472A - System for processing data base - Google Patents

Abstract

PURPOSE:To shorten the waiting time of another access against a disk device during the processing of data base concerning a disk controller. CONSTITUTION:When the data base processing is instructed in the disk controller 3 provided with a data base processing function 30 where data being necessary for a master device 27 is extracted and transmitted, the processing instruction of data quantity designated by a data division table 28 is sequentially recorded in a task control table 29 by a previously fixed time interval. During this, the data base processing function 30 sequentially processes data of designated quantity by the recorded order of the task control table 29, an input/output instruction received during data base processing is recorded by interrupting to the task control table 29 at the point of time of receiving and the recorded input/output instruction is executed in the recorded order.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk control device for controlling data transfer between a disk device and a host device, and in particular, to
The present invention relates to a database processing method for reducing the waiting time for other accesses to a disk device during database processing for selecting records requested by a host device.

[0002] In recent years, in order to improve the efficiency of computer systems, only records needed by the central processing unit are selected in the disk control unit from among the data to be processed stored in the disk unit, and transferred to a higher-level unit such as a channel. Database Assist (Data Bass)
Demand for e-assistants (hereinafter abbreviated as DBAs) is increasing.

[0003] Generally, data is stored in a disk device in units of pages, one page of data is made up of a plurality of records, and one record is made up of a plurality of columns.

[0004] For example, if each record is made up of multiple columns such as name, age, and address, and a record of a certain age group is to be extracted and processed, the disk control device If the age group column is collated and the central processing unit extracts and transfers only the records of the required age group, the higher-level device does not need to receive transfers of unnecessary records, reducing data transfer time. At the same time, the processing efficiency of the computer system is improved because the time for selecting records is saved.

However, when performing such database processing on a disk control device equipped with a DBA, the DBA transfers a large amount of data from the disk device to the memory.
Although the disk device is occupied for a long time for processing,
During this time, it is necessary to ensure that other accesses to the disk device are not made to wait.

[0006]

2. Description of the Related Art FIG. 6 is a block diagram illustrating an example of the prior art, and FIG. 7 is a detailed block diagram of each functional unit.

The disk control device 3 is composed of a plurality of functional units, and the channel adapters 5 and 6 are connected to each other as shown in FIG.
), the processor 15 has a control memory 16
reads the program stored in the , and receives the start I/O command from channel 1 or 2 through the interface circuit 13 and transfers data to and from channel 1 or 2, as well as the common bus control circuit. 14 by controlling
Data is transferred to and from other functional units via the common bus 12.

Device adapters 7 and 8 are shown in FIG. 7(A).
With the configuration shown in FIG. 1, the processor 15 reads a program stored in the control memory 16 and operates, sends commands to the disk device 4 via the interface circuit 13, and transfers data to and from the disk device 4. , controls the common bus control circuit 14 to transfer data to and from other functional units via the common bus 12.

The DBA 9 has a configuration as shown in FIG. Then, an instruction is sent to the device adapter 7 or 8 from the channel 1 or 2 via the channel adapter 5 or 6 to read data in the specified address range of the disk device 4 to the shared memory 11 .

When data is stored in the shared memory 11, the data is read out via the common bus control circuit 21, stored in the RAM 17, and sent to the comparison circuit 18 for channel 1.
or 2 instructs to select a record having the same column as the designated column, and the comparison circuit 18 reads data from the RAM 17 record by record, and stores the selected record in the RAM 17 when it selects a record that has the same column as the designated column.

When the processor 19 completes the selection of data in the specified address range from channel 1 or 2,
The record selected in the RAM 17 is stored in the shared memory 11.

[0012] The resource manager 10 is shown in FIG. 7(C).
With the configuration shown in FIG. 1, the processor 23 reads the program stored in the control memory 22 and operates, controls the common bus control circuit 24, and controls all starts from channel 1 or 2 via the common bus 12. Receive I/O command and R
The start I/O command is stored in the AM 25, and the start I/O command is centrally managed, and data transfer processing instructions are given to the channel adapters 5, 6 and device adapters 7, 8 for each start I/O command.

[0013] Furthermore, control information for each received start I/O command, control information for each channel path, and control information for each received start I/O command,
control information is stored in the RAM 25, and other functional units are
This RAM 25 is accessed via the M access control circuit 26, and the above information can be read out, thereby centrally managing this information.

For example, channel 1 is connected to channel adapter 5.
A start I/O command is sent to the disk device 4.
When the channel adapter 5 specifies the address of the resource manager 10, that is, the cylinder address, head address, and record address, and instructs to write data, the channel adapter 5
If permission is granted, the address of the device adapter 7 is sent, for example, and the connection is made.

When this connection is completed, the channel adapter 5 sends data from the channel 1 and stores the data in the shared memory 11. The device adapter 7 is
The data stored in the shared memory 11 is read out, sent to the disk device 4, and written to a designated address.

When instructed to read data from the channel 1, the channel adapter 5 causes the connected device adapter 7 to read the data from the specified address of the disk device 4, and stores the data stored in the shared memory 11. is read and transferred to channel 1.

When the channel adapter 6 is instructed to perform database processing from the channel 2, the channel adapter 6
It requests the resource manager 10 to use the common bus 12, and if granted, sends and combines the address of the DBA 9, and notifies the DBA 9 of the specified address range of the disk device 4.

Upon obtaining permission from the resource manager 10, the DBA 9 sends and combines the addresses of the device adapters 8 via the common bus 12, and instructs to read data in the specified address range from the disk device 4. .

Therefore, the device adapter 8 occupies the disk device 4, reads data within the designated address range, and stores it in the shared memory 11. As mentioned above, when data starts to be stored in the shared memory 11, the DBA 9
When this data is fetched, a record having the same column as the designated column is selected, and the record selected from the data within the designated address range is stored in the shared memory 11, the resource manager 10 is notified of the end.

The resource manager 10 remembers that the channel adapter 6 has been instructed to perform database processing, and instructs the channel adapter 6 to send the record stored in the shared memory 11 to the channel 2.

[0021]

FIG. 8 is a diagram illustrating the problems of the prior art. As shown in the database processing in FIG. 8, if the address range of the disk device 4 specified by channel 2 is large, the device adapter 8
This occupies the disk device 4 and increases the time it takes to transfer data to the shared memory 11.

Therefore, during the time shown in this database processing, channel 1 or 2 is used as shown by the dotted line in the general processing.
Since the device adapter 8 occupies the disk device 4, there is a problem in that the general input/output command sent by the device adapter 8 becomes busy and has to wait for a long time from the time the input/output command is sent.

In view of these problems, the present invention does not perform database processing continuously, but divides it into several time periods, and performs general operations between the divided time periods. The purpose is to reduce the waiting time for general input/output commands by making it possible to interrupt input/output commands.

[0024]

Means for Solving the Problems FIG. 1 is a block diagram illustrating the principle of the present invention. The disk control device 3 controls data transfer between the host device 27 and the disk device 4, and also extracts data required by the host device 27 from data within the address range of the disk device 4 specified by the host device 27. It is equipped with a database processing function 30 that extracts and transfers data.

[0025] Also, a data partition table is provided that specifies the amount of data to be read from the disk device 4 during database processing in accordance with a predetermined partition amount based on the frequency of receiving input/output commands other than commands instructing database processing. 28, and a record that instructs execution of database processing at predetermined time intervals based on the amount of data specified by this data division table 28, and a record that instructs execution of an input/output command at the time it is received. A task control table 29 for performing the following is provided.

When database processing is instructed by the host device 27, the data division table 28 is added to the task control table 29 at the predetermined time interval.
While sequentially recording processing instructions for the amount of data specified by The received input/output command is stored in the task control table 29 at the time it is received.
The recorded input/output instructions are also executed in the recorded order.

[0027]

[Operation] With the above configuration, the processing of the amount of data specified by the data division table 28 is sequentially recorded at predetermined time intervals in the task control table 29 that instructs the operation procedure of the disk control device 3. , the input/output commands are interrupted and recorded at the time of reception, so the database processing and the input/output commands are executed in the order recorded in the task control table 29.

[0028] Therefore, the time required for execution of input/output instructions to be waited can be reduced.

[0029]

[Embodiment] FIG. 2 is a block diagram of a circuit showing an embodiment of the present invention, and FIG. 3 is a detailed block diagram of a functional unit of the present invention.
FIG. 4 is a diagram illustrating an example of a table.

The same reference numerals as in FIGS. 6 and 7 indicate the same functions. The DBA 31 in FIG. 2 has a configuration as shown in FIG. 3(A), and the resource manager 32 in FIG. 2 has a configuration as shown in FIG. 3(B).

When the DBA 31 is instructed to perform database processing by specifying the address range of the disk device 4 from the channel adapter 5 or 6, the processor 19 receives an input other than an instruction for database processing stored in the control memory 33. This address range is divided according to predetermined conditions based on the frequency of receiving output commands, and a data division table 28 is created.

That is, the specified address range is, for example, if the cylinder number is in hexadecimal notation "0000" to "
0100", and the head number is "000" in hexadecimal.
0'' to ``000E'' and the record number is ``01'' to ``1F'' in hexadecimal notation, the range of database processing at one time is divided into, for example, 5 tracks each, and the data division shown in Figure 4 is performed. Create a table that specifies the range to be processed, as shown in tables (■) to (■).

[0033] That is, ■ is the cylinder numbered "0000",
The head numbered "0000" starts reading data from the record numbered "01", and after reading up to the record numbered "1F", the head is sequentially read from the numbered records "0001" to "000".
4” and then switch from record “01” to “1F”.
The data is read up to the record numbered “0000” and the head numbered “0004” is used to read the data to “1F”.
” indicates that reading ends when the record numbered is read.

[0034] And the last row of the table is "0100"
With the cylinder numbered “000A” and the head numbered “000A”,
Read data from record number 01” and set it to “0100”
With cylinder No. 000E and head No. 000E,
When the record number 1F'' has been read, this indicates that data reading from the disk device 4 is complete.

[0035] The processor 19 uses the data partition table 2
8 is created, the resource manager 32 is sent sequentially from Notify the divided processing range.

Processor 23 of resource manager 32
creates a task control table 29 in the RAM 34 based on this processing target range. That is, as shown in the task control table of FIG. 4, TCB1 indicating the task control block
-Create TCBn sequentially.

That is, when the DBA 31 instructs execution of processing based on the processing target range sent out at predetermined time intervals and an input/output command is sent out from the channel adapter 5 or 6, the time when this input/output command is received. Then, create a task control table that instructs the execution of this input/output command.

Therefore, for example, as shown in the figure, TCB (
1) becomes "Execute staging of the address range specified by ■", TCB (2) becomes "Execute staging of the address range specified by ■", and TCB
(3) becomes "Execute the input/output command", and TCB (4
) becomes "Execute staging of the address range specified by ■."

The processor 23 uses the task control table 29
are created sequentially, and the contents sequentially instructed from TCB (1) of this task control table 29 are executed in each functional unit.

That is, TCB (1), (2) and (4)
For example, if the device adapter 8 executes the TCB
(3) is executed by the channel adapter 5 and device adapter 7. FIG. 5 is a diagram illustrating the operation of FIG. 2.

As shown in the database processing in FIG. 5, the device adapter 8 connects TCBs (1), (2) and (4).
etc., the time required to occupy the common bus 12 and the disk device 4 and transfer data to the shared memory 11 is short, and after the completion of processing TCB (2), while executing TCB (4), the general As shown in the process, input/output commands can be executed.

[0042] Therefore, the waiting time for general processing can be shortened.

[0043]

As explained above, since the present invention performs database processing, the processing of general input/output commands does not have to wait for a long time, and when multiple database processing is requested, Parallel processing can be performed by recording other database processing items instead of executing input/output commands in the task control table.

[Brief explanation of the drawing]

[Figure 1] Block diagram explaining the principle of the present invention

[Figure 2
] Block diagram of a circuit showing one embodiment of the present invention

[Figure 3
] Detailed block diagram of functional units of the present invention

[Figure 4]
Diagram explaining an example of a table

[Figure 5] Diagram explaining the operation of Figure 2

[Figure 6] Block diagram explaining an example of conventional technology

[Figure 7] Detailed block diagram of each functional unit

[Figure 8] Diagram explaining the problems of the conventional technology

[Explanation of symbols] 1, 2 Channel 3 Disk control device 4 Disk device 5, 6 Channel adapter 7, 8 Device adapter 9, 31 DBA 10, 32 Resource manager 11 Shared memory 12 Common bus 13, interface circuit 14, 21, 24 Common bus control circuit 15, 19, 2
3 Processor 16, 20, 22, 33 Control memory 17, 25, 34 RAM 18 Comparison circuit 26 RAM access control circuit 27 Host device 28 Data division table 29 Task control table 30 Database processing function

Claims (1)

[Claims] [Claim 1] A host device (27) and a disk device (4
), as well as controlling data transfer between the host device (
A disk control device (3) equipped with a database processing function (30) that extracts and transfers data required by the host device (27) from data within the address range of the disk device (4) specified by the device (27). a data partitioning table (28) that specifies the amount of data to be read from the disk device during database processing in accordance with a predetermined partitioning amount based on the frequency of receiving input/output commands other than commands instructing database processing; Based on the amount of data specified by the data division table (28), records are made to instruct execution of database processing at predetermined time intervals, and instructions are given to execute the input/output command when received. A task control table (29) for recording data is provided, and when database processing is instructed, the data division table (28) specifies the data at the predetermined time interval in the task control table (29). While sequentially recording processing instructions for the amount of data, the database processing function (30) is caused to sequentially process the designated amount of data in the order of recording in the task control table (29), and the input information received during the database processing is A database processing method characterized in that the output command is recorded by interrupting the task control table (29) at the time of reception, and the recorded input/output commands are also executed in the recording order.