JPH01300479A - System for diagnosing patrol of magnetic disk device - Google Patents

Abstract

PURPOSE:To prevent the performance of an I/O processing time from being lowered by evading a seek operation for a magnetic disk device with I/O start-up after the lapse of a constant time, and restarting the seek operation if no I/O start-up is performed by a time when the constant time elapses further. CONSTITUTION:In the initializing state of a patrol diagnosing device 11, all of the bit maps 14 are set at 1s, and after the constant time is counted by a timer 12, the value of a counter 13 is increased by 1, and it is returned to 0 when a value on which 1 is added exceeds the maximum connecting number (n) of the magnetic disk devices (0-n-1). When the bit of the bit map 14 corresponding to the value of the counter 13 is 1, a cylinder address on which a magnetic head is located is read from the magnetic disk device corresponding to the value of the counter 13 (the magnetic device 0 is selected when the value of the counter is 0, and so forth). And a cylinder address to be sought next is found by a known calculation equation, and seek is started up on the cylinder. In such a way, it is possible to prevent the performance of the I/O processing time from being lowered.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a magnetic disk control device, and more particularly to patrol diagnosis in which the magnetic head of a magnetic disk device to be connected is moved at regular intervals.

BACKGROUND OF THE INVENTION In recent years, technological advances in magnetic disk drives have been remarkable, and smaller, larger-capacity magnetic disk drives have been developed. As the recording bit density and track density of magnetic disk media increase, it is necessary to reduce the distance between the medium surface and the magnetic head surface, and this technology is one of the elements necessary for miniaturization and large capacity.

On the other hand, as the distance between the medium surface and the magnetic head surface becomes smaller, it is necessary to remove as much dust as possible from the medium surface and the vicinity of the magnetic head. In particular, in situations where the magnetic head does not move for a long time and remains in a fixed position, dust may adhere between the media surface and the magnetic head, causing damage to the media surface and the magnetic head. Become. A patrol diagnosis has been proposed in which the magnetic head is moved by a predetermined amount once every certain period of time to prevent dust from adhering between the medium surface and the magnetic head.

Problems to be Solved by the Invention In the conventional patrol diagnosis described above, a magnetic disk control device that connects and controls magnetic disk devices measures a certain period of time, and when a certain period of time has elapsed, one of the subordinate magnetic disk devices is activated. The head was moved relative to the

However, when starting I/O from a host device to a magnetic disk device, software accesses the stored data after positioning the desired data on the disk where it is stored. Since the patrol diagnosis automatically changes the position of the magnetic head, an extra return operation of the magnetic head is required in order for the software to access the desired stored data.

This extra return operation reduces performance in I and /O processing time.

The present invention has been made in view of the above-mentioned conventional circumstances, and therefore, an object of the present invention is to provide a new roll diagnosis method for a magnetic tissue device that makes it possible to solve the above-mentioned problems inherent in the conventional technology. Our goal is to provide the following.

Means for Solving the Problems In order to achieve the above object, the B)~roll diagnosis method according to the present invention includes a timer that measures a certain period of time, and a timer that increments by 1 every time the timer measures a certain period of time. A counter that cyclically sets the maximum number of magnetic disk devices to be connected and controlled by 1, and a bit corresponding to the magnetic disk device that was started when an I/O of the host device was started is set to O, and the above-mentioned Each magnetic disk device has a map corresponding to a bit, which updates a bit corresponding to the value of the counter incremented by 1 to 1 when the timer measures a certain period of time;
The value indicated by the counter when the corresponding bitmap is already 1 before the bitmap corresponding to the value indicated by the counter is incremented by 1 when the timer measures a certain period of time. Embodiment Next, a preferred embodiment of the present invention will be described. Examples will be specifically described with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention. Referring to FIG. 1, an embodiment of the patrol diagnosis method according to the present invention is realized by a patrol diagnosis device 11, and this patrol diagnosis device 11 includes a timer 12,
Counter 13 and Bitmuff.

14 and seek means 15, and connects and controls the magnetic disk drives 0 to (n-1).

FIG. 2 shows the operation of the one-roll diagnostic apparatus shown in FIG. 1 in the form of flowchart 1.

=5- In the initial state of the patrol diagnostic device 11, all bitmaps 14 are set to 1. The timer 12 measures a certain period of time. The fixed time is the interval at which seek is activated in the magnetic disk device, and is determined by the maximum number of connected magnetic disk devices and the number of cylinders, but since this is not important for a detailed explanation of the present invention, the details will be omitted. .

When the timer 11 measures a certain period of time, the value of the counter 12 is incremented by 1. If the +1 value exceeds the maximum number n of connected magnetic disk devices, it is returned to 0.4If the bit of the bit map 14 corresponding to the value of the counter 13 is 1, the magnetic disk device corresponding to the value of the counter 13 (the value of the counter If is O, read the cylinder address where the magnetic head is located from the magnetic disk drive 0 (the same applies hereafter), use a known calculation formula to find the cylinder address to which the primary seek should be performed, and start a seek to this cylinder. do.

The calculation formula is determined by the maximum number of connected units n, the number of cylinders, etc., but the details are omitted because they are not important for the detailed explanation of the present invention. .

When there is an I/O activation from a host device, the bitmap 14 corresponding to the magnetic disk on which the I/O activation occurred is set to 0, and then the I/O processing is performed.After this, a certain period of time elapses. , if the magnetic disk device corresponding to the value after incrementing the counter 13 by 1 is the magnetic disk device in which the I/O was started earlier, the bit of the corresponding bitmap 14 is set to O, so the above-mentioned seek cannot be performed. Avoid startup, set the bit of bitmap 14 to 1, and return to the original state.

As described in detail, according to the present invention, a seek operation is performed once every fixed period for a magnetic disk device without I/O activation, but a seek operation is performed for a magnetic disk device with I/O activation. In contrast, a seek operation after a certain period of time is avoided. After that, if there is no I/O activation within a certain period of time, the seek operation will be restarted.3 Therefore, a seek operation will be performed for a magnetic disk device that is not accessed by software, but from software. For magnetic disk drives that are frequently accessed, in order to avoid seek operations, patrol diagnosis can be performed to prevent dust from entering the magnetic head and media surface without causing a performance drop in I/O processing time. This has the effect of preventing adhesion between the parts.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
The figure is a flowchart showing the operation of the present invention. 11...Patrol diagnosis device, 12...Timer, 13
...Counter, 14...Bitmap, 15...Seek means Patent applicant: NEC Corporation Representative, Patent attorney Yutabe Kumagai

Claims (1)

[Claims] In a magnetic disk control device that connects and controls multiple magnetic disk devices, the maximum number of magnetic disk devices to be connected and controlled is determined by using a timer that measures a certain period of time and adding one each time the timer measures a certain period of time. The counter that counts each click has an initial value of 1, and the bit corresponding to the magnetic disk that was started when I/O was started from the host device is set to 0, and when the above-mentioned timer measures a certain amount of time, the bit is increased by 1. A bit map for updating the bit corresponding to the value indicated by the counter to 1 is provided for each magnetic disk device, and when the timer measures a certain period of time, the bit map is updated to 1.
When the bitmap corresponding to the value indicated by the counter is updated to 1, the magnetic head of the magnetic disk device corresponding to the value indicated by the counter is updated to 1. 1. A patrol diagnosis method for a magnetic disk drive, comprising a seek means for causing a cylinder to seek a cylinder determined by a known calculation formula for a cylinder in which the cylinder is located.