JPS61221842A - Computer system - Google Patents

Abstract

PURPOSE:To attain the integrated handling of programs by giving a function to a disk device controller to control these disk devices after allocating the capacity of each disk device to the continuous memory spaces. CONSTITUTION:A host system sends a desired logical address to a disk controller for delivery of an indication. The comparison is performed within the disk controller among a logical address LA received from the host system and the first and last logical addresses SLA and ELA which are occupied by the disk devices registered previously into a disk memory mapping table. Thus the corresponding disk number is obtained. Then the cylinder number SI, the sector number SE per track and the head number H are obtained from said decided disk number. Based on these numbers SI, SE and H and the addresses LA and SLA, the cylinder number, the sector number and the head number are obtained for the disk device to which an access is actually given and delivered to the disk device.

Description

[Detailed Description of the Invention] Technical Field> The present invention relates to a disk device controller in a computer system in which a plurality of disk devices are connected. Computer·
In the system, the host system managed things like switching disk devices and allocating memory space. For example, in applications that handle a large amount of data that exceeds the capacity of one disk device, the host system allocates the data that exceeds the capacity of one disk device to another disk device. Normally, when writing data to or reading data from a disk device, it is necessary to specify the disk number (drive number), cylinder number, sector number, head number, etc. In the case of data spanning two disk devices, it was necessary to specify them twice. For example, when creating a program using a disk device, if the data to be used exceeds the capacity of a single disk device, or if the disk device is already occupied for other purposes and there is little remaining capacity. Programs cannot be executed with only one disk device. Naturally, in a system having multiple disk devices, a second disk device must be used. For this reason, exceptional processing that deviates from the original purpose of the program is required, such as switching to a second disk device when the capacity of the first disk device is exceeded.

As a result, there was a problem in that programs could not be handled in a unified manner, resulting in a lack of flexibility.

<Objective and Structure of the Invention> The present invention has been made with the aim of solving the above-mentioned problems.In a system in which a plurality of disk devices are connected, the capacity of each disk device is divided into a continuous memory space. The disk device controller has the function of allocating and managing disk drives.

<Effects of the Invention> With the above configuration, the host system no longer needs to consider the capacity of each individual disk device, and can treat multiple disk devices as one virtual large-capacity disk device. can be done. As a result, the burden on the host system is reduced, and even when creating an application program, disk devices can be handled in a uniform manner, increasing the generality of the program.

<Embodiment> FIG. 3 shows a system configuration diagram, and FIG. 4 shows a memory space of a virtual disk device. In Figure 3, 1 is the host system, 2 is the disk controller, and 3 o
+31+..., 3N are magnetic disk devices.

Actual disk controller functions include the following.

■ Change read and write commands issued by the host system to virtual disk devices into actual commands to individual disk devices.

■ Allocate the capacity of individual disk units into contiguous memory space.

These features also allow additional disk devices to be connected without changing the host system.

This will be explained below using an example.

FIG. 1 is a block diagram showing the configuration of a disk controller according to the present invention.

Each block will be explained.

・Host interface block 11 host・
System l and logical address LA.

Supports Data D's Yashi and Shi.

- Disk interface block 12 supports control signals of disk device 3.

- Format control block 13 Supports writing/reading of data to/from the disk device 3 (data signals, etc.).

・Processor block 14 Controls the inside of the disk controller.

Converts a logical address to a physical address.

- ROM15 A program for controlling the processor block is stored.

- RAM 16 Stores the disk memory mapping table of the disk device. The disk memory mapping table is used in the present invention to convert a logical address into a physical address (disk number, cylinder number, sector number, head number) of each disk device.

An example of the disk memory mapping table is shown in the second example.
As shown in the figure.

Note that the contents of this RAM, ie, the disk memory mapping table, are initialized (written to the RAM) when the system is started.

Internal bus 17 is used for data transfer within the disk controller.

In FIG. 1, when the disk device is accessed, the following processing is performed.

■ The host system instructs the disk controller by sending the desired logical address.

■ Inside the controller, the logical address (LA) received from the host system, the first logical address (SLA) and the last logical address occupied by each disk unit in the disk memory mapping table registered in advance are stored. The address (ELA) is compared to determine the disk number (drive number) to be lost. That is, S.L.A.
Search for a disk number (drive number) that satisfies the formula ≦LA≦ELA.

■ From the determined disk number (drive number), the items in the mapping table at that time, that is, the number of cylinders (s) of the disk device, the number of sectors per track (
SE), Hera)' number (H) is found.

■ Required+'cS1. SE, H! LA, SLA
From this, the cylinder number No. (SI'), sector number No. (sE), and head number No. (H) of the disk device to be actually accessed are determined. However, the logical address is in units of sectors.

-8IXNo(H) Sector number=No(SE)=LA-8LA-8E(IN
T CXJ represents the integer part of x. ) ■ Output the disk number (drive number), cylinder number, sector number, and head number obtained above to the disk device.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a disk controller according to the present invention, and FIG. 2 is a block diagram showing the configuration of a disk controller according to the present invention.
A diagram showing an example of a table, FIG. 3 is a system configuration diagram of a computer system according to the present invention, and FIG. 4 is a memory space diagram of a virtual disk device in the same system. Explanation of symbols 1: Host system, 2: Disk controller,
3": Disk device, 3o + 31+ "'+ 3N
m magnetic disk device, 11: host interface block, 12: disk interface block;
Block, ) 3 near-automatic control block, 14゛:
Processor Φ block, 15: ROM. 16: RAM, 17: Internal bus, LA: Logical distribution, D
:data. Agent: Patent attorney Aihiko Fukushi (and 2 others), ≧T,
In September 175f<27・Ko,ky-fuqjaAu
``f Pro Noah Figure 1 Roz 7'' Memory ``Mabbinde Dry Bu'' Lurie Δ? h,
+, if. Figure 2

Claims (1)

[Claims] 1. A computer system in which a plurality of disk devices are connected is provided with a disk device controller that allows the host system to handle the capacity of each of the disk devices as a continuous memory space. computer system.