公开(公告)号：JPH1139783A

公开(公告)日：1999-02-12

申请号：JP18790097

申请日：1997-07-14

Applicant: IBM

Inventor： OKUTSU KAZUE ,  UEDA TETSUO ,  FUKUSHIMA YUKIO ,  KAKIHARA TOSHIO

IPC: G11B21/08 ,  G06F12/00 ,  G11B19/20

Abstract: PROBLEM TO BE SOLVED: To provide the start-up method for the information recorder to use a computer system as quickly as possible. SOLUTION: When the information recorder equipped with an actuator arm for supporting a flexible cable is to be started up by this method, first of all, a power source is supplied (a process 201). Then, a start-up operation excluding production of a correcting tension data of the flexible cable is executed (a process 202). Afterward, an access command from a host computer is made to be in the executable state (processes 203 and 204). Subsequently, the production of the correcting tension data of the flexible cable is executed (a process 205). Consequently, the command can be processed prior to the production of the correcting tension data.

公开(公告)号：JPH1173279A

公开(公告)日：1999-03-16

申请号：JP22968797

申请日：1997-08-26

Applicant: IBM

Inventor： KAKIHARA TOSHIO ,  UEDA TETSUO ,  OKUTSU KAZUE ,  FUKUSHIMA YUKIO

IPC: G06F3/06 ,  G06F12/08 ,  G11B20/10 ,  G11B20/18 ,  G11B19/04

Abstract: PROBLEM TO BE SOLVED: To evade an increase in the overhead of a command by determining whether or not the depth of a queue is decreased when an error is detected if cached data are written to a rotary data recording disk. SOLUTION: Data stored in N (N; integer

公开(公告)号：JP2003316523A

公开(公告)日：2003-11-07

申请号：JP2002110702

申请日：2002-04-12

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーション

Inventor： NISHINOMIYA HIROMI ,  KAKIHARA TOSHIO ,  YUZUBA AKINORI ,  OYAMA TOMOYUKI

IPC: G06F3/06 ,  G06F12/00 ,  G11B20/10

CPC classification number: G06F3/0659 ,  G06F3/061 ,  G06F3/0674

Abstract: PROBLEM TO BE SOLVED: To provide a method for selecting an execution command capable of improving performance when the depth of a queue is deep.
SOLUTION: An execution standby time D1 of a command C1 exceeds the 1/4 cycle of a reference time CRT. An execution standby time D8 of a command C8 is 8/64 (1/4) cycle, and this fulfills requirements superior to reference time CRT. Also, the execution standby time D8 of the command C8 is not less than a scanning time SCN. Therefore, the command C8 is selected as a command to be selected the next. After this command is selected, the calculation of the execution standby time for any following command posterior to a command C9 is stopped.
COPYRIGHT: (C)2004,JPO

Abstract translation: 要解决的问题：提供一种当队列深度较深时能够提高性能的执行命令的选择方法。 解决方案：命令C1的执行待机时间D1超过参考时间CRT的1/4周期。 命令C8的执行待机时间D8为8/64（1/4）周期，这满足了优于参考时间CRT的要求。 此外，命令C8的执行待机时间D8不小于扫描时间SCN。 因此，命令C8被选择为下一个选择的命令。 选择此命令后，停止对命令C9后面的任何后续命令的执行待机时间的计算。 版权所有（C）2004，JPO

公开(公告)号：JPH1145157A

公开(公告)日：1999-02-16

申请号：JP19872897

申请日：1997-07-24

Applicant: IBM

Inventor： SATO MASAHIKO ,  KAKIHARA TOSHIO ,  KANAMARU ATSUSHI ,  OSHIKAWA HIROSHI ,  ASANO HIDEO

IPC: G06F3/06 ,  G06F13/38 ,  G11B20/10 ,  G11B27/10

Abstract: PROBLEM TO BE SOLVED: To permit high-reliability data transfer and a minimum throughput decrease by controlling data transfer to a 2nd data transfer speed slower than a 1st data transfer speed once an error detecting means detects a specific error. SOLUTION: A hard disk drive(HDD) is equipped with a hard disk controller 105A which controls operation for reading and writing data out of and to a magnetic disk 101 and an MPU 106 which controls the operation of the whole HDD. To transfer data from the HDD 100 to a host 200, the 1st data transfer speed is set first and the data are transferred from the HDD to the host 200 at the 1st data transfer speed. If a specific error is detected in the transferred data, the data transfer speed is changed to the 2nd transfer speed slower than the 1st transfer speed. Consequently, the state of the IDEI/FBus can be known by monitoring the status of the existent CRC error, so high-reliability data transfer for reading can be performed only by alterations of the drive side.

公开(公告)号：JPH10275110A

公开(公告)日：1998-10-13

申请号：JP7494997

申请日：1997-03-27

Applicant: IBM

Inventor： KANAMARU ATSUSHI ,  KAKIHARA TOSHIO ,  KIGAMI YUJI ,  SAITO TAKAHIRO

IPC: G06F3/06 ,  G06F12/08 ,  G11B19/02

Abstract: PROBLEM TO BE SOLVED: To provide the disk drive device and its control method which can make an additional read of media at need while minimizing a command overhead, and then improve the total performance. SOLUTION: The disk drive device 10 performs data transfer until final one block is left and expects the intervention of the local MPU 16 even when HIC 15 is able to perform all data transfer without the intervention of a local MPU 16 while a read command is issued and requested data hit data held in a cache memory, the local MPU 16 makes an indication for transferring the final one block once preparations for the end of the command are made, and if there are more than one block of transfer data left when the local MPU 16 instructs the HIC 15 to transfer the final one block, the HIC 15 performs all the data transfer without interrupting the data transfer at the final one block, thus completing the command.

公开(公告)号：JPH10255403A

公开(公告)日：1998-09-25

申请号：JP5775197

申请日：1997-03-12

Applicant: IBM

Inventor： KAKIHARA TOSHIO ,  KANAMARU ATSUSHI ,  SATO TAKUMI ,  SATO MASAHIKO

IPC: G06F3/06 ,  G11B19/04 ,  G11B20/12 ,  G11B20/18 ,  G11B27/36

Abstract: PROBLEM TO BE SOLVED: To reduce the error processing time during a reset, by executing a first error recovering processing while controlling a write controlling means if a writing error is detected and executing a second error recovering processing, which is small in the number of execution commands than an error recovering processing(ERP), while controlling the means if a reset instruction is detected. SOLUTION: If a reset is requested, a control section 3 does not execute the ERP, the selector being selected is assumed to be defective, the temporary registration processing of the defective sector is conducted, the assignment of a spare sector is made and the updating of a reassigned table is executed. In the normal ERP having no reset request, many processings are repeated from such a judgement that is not a reset, up to the judgement of an error recovering impossible. Thus, the number of steps, which are actually to be executed, becomes extremely large. Therefore, the time required to execute the processings greatly differs for between the case of a reset is requested and the case in which no request is made for a reset.

公开(公告)号：JPH1145524A

公开(公告)日：1999-02-16

申请号：JP19554597

申请日：1997-07-22

Applicant: IBM

Inventor： FUKUSHIMA YUKIO ,  UEDA TETSUO ,  KAKIHARA TOSHIO ,  ASANO HIDEO

IPC: G11B21/10 ,  G11B5/596

Abstract: PROBLEM TO BE SOLVED: To prevent an erroneous correcting operation and to suppress the deterioration of performance by detecting a flaw in one of a pair of burst patterns or more recorded in each of plural tracks and positioning a head in a prescribed track in accordance with a position error signal generated by using the other burst pattern. SOLUTION: The disk device 30 is wholly controlled by an MPU 42 together with a hard disk controller HDC 40 in accordance with a control program. Plural tracks, each comprising a data area for recording data and a servo area in which a pair of burst patterns or more are recorded in advance, are disposed in a pattern of concentric circles in the surface of a disk 31. When a read/write error due to a flaw in one burst pattern read out by a magnetic head 33 is formed, this burst pattern is not used but the other one is used to generate a positioning error signal by the MPU 42. The magnetic head 33 is positioned into the center of a prescribed track by the MPU 42 in accordance with this positioning error signal. Consequently, the right correcting operation can thus be performed.

公开(公告)号：JPH10275111A

公开(公告)日：1998-10-13

申请号：JP7513997

申请日：1997-03-27

Applicant: IBM

Inventor： KANAMARU ATSUSHI ,  KAKIHARA TOSHIO ,  ASANO HIDEO ,  TOBARI ATSUSHI

IPC: G06F3/06 ,  G06F12/08 ,  G11B19/02

Abstract: PROBLEM TO BE SOLVED: To provide the disk drive device and its control method which can reduce writing to media while minimizing a command overhead, and then improve the total performance. SOLUTION: The disk drive device 10 has HDC13, a cache memory 14, and a command queue holding commands cached in the cache memory 14 and is equipped with HIC 15 which processes host interface peripheries through hardware and a local MPU 16 which includes control over the HDC 13 and HDC 15 and controls the whole HDD; and the device indicates the writing of data to media to the HIC 15 with a write command cached in the cache memory 14, retrieves a retrieval source command which is completely overwritten with the latest issued command among write commands cached in the cache memory 14, and does not write data to the media with the overwritten retrieval source command.

公开(公告)号：MY118820A

公开(公告)日：2005-01-31

申请号：MYPI9800731

申请日：1998-02-20

Applicant: IBM

Inventor： KANAMARU ATSUSHI ,  KAKIHARA TOSHIO ,  KIGAMI YUHJI ,  SAITOH TAKAHIRO

IPC: G06F12/08 ,  G06F3/06

Abstract: A DISK DRIVE AND A METHOD FOR CONTROLLING THE DISK DRIVE IN WHICH AN ADDITIONAL READ FROM A MEDIUM CAN BE PERFORMED AS NEEDED WHILE MINIMIZING THE COMMAND OVERHEAD TO THE MINIMUM. IN A DISK DRIVE DEVICE (10), EVEN IF A READ COMMAND IS ISSUED, AND ALL THE REQUIRED DATA IS STORED IN THE CACHE MEMORY SUCH THAT THE DATA TRANSFER CAN BE PERFORMED WITHOUT THE INTERVENTION OF A LOCAL MPU (16), A HIC (15) PERFORMS THE DATA TRANSFER SO AS TO LEAVE THE LAST ONE BLOCK AND WAITS FOR THE INTERVENTION OF THE LOCAL MPU (16). THE LOCAL MPU (16) PROVIDES INSTRUCTIONS TO TRANSFER THE LAST ONE BLOCK WHEN THE PREPARATION FOR THE COMMAND TERMINATION IS COMPLETED, AND IF A PLURALITY OF BLOCKS OF DATA TO BE TRANSFERRED ARE REMAINING WHEN THE LOCAL MPU (16) INSTRUCTS THE HIC (15) TO TRANSFER THE LAST ONE BLOCK, THE HIC (15) EXECUTES ALL THE DATA TRANSFER WITHOUT STOPPING THE DATA TRANSFER BEFORE THE LAST ONE BLOCK, COMPLETING THE COMMAND.FIGURE 1