Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A handling method of read commands which hit the cached data located on the separated segments.

IP.com Disclosure Number: IPCOM000010430D
Original Publication Date: 2002-Nov-29
Included in the Prior Art Database: 2002-Nov-29
Document File: 2 page(s) / 19K

Publishing Venue

IBM

Abstract

Disclosed is a method to improve drive read performance in AT HDD .It is a handling method of read commands which hit the cached data located on the separated segments. The HDD has not to read the data from disk when the data exist two or more segments separately.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 50% of the total text.

Page 1 of 2

Invention Disclosure Publication for JP8-2002-0768

Created By Akihiko Nadaguchi On 2002/10/22 16:14:45

Last Modified By Chieko Kobayashi On 2002/11/29 11:11:59

Required fields are marked with the asterisk ( * ) and must be filled in to complete the form .

Editors of this article Shuhji Yamada/Japan/IBM, Hiroshi Uchiike/Japan/IBM, Takahiro Saito/Japan/IBM Status Published
Title A handling method of read commands which hit the cached data located on the separated segments. Inventors Shuhji Yamada, Hiroshi Uchiike, Takahiro Saito
Attorney Akihiko Nadaguchi/Japan/IBM
Functional Area (R80,R90 STD F/I, LSI, C/I) R80,R90 STD F/I, LSI, C/I
click on "Select" in the action bar
Processing Loc JP

Publication Information

Date Approved

Publication Name Research Disclosure Date Sent 2002/11/29

Date Published 2003/01

Article Number Page Number Issue Number Number of Pages

* Enter a brief abstract for this article

Disclosed is a method to improve drive read performance in AT HDD .It is a handling method of read commands which hit the cached data located on the separated segments. The HDD has not to read the data from disk when the data exist two or more segments separately.

    * Enter the body of your Invention Disclosure Publication below. You may include drawings and highlighting as appropriate. HDD has DRAM called Sector Buffer. The Sector Buffer keeps some data to improve read  performance. The series of stored data are called as a segment. HDD has a Segment Information  Table to manage the segments in Sector Buffer.

--- This is a example of Segment Information Table ---

Segment ID | Last LBA | Length | LastPage -----------+----------+--------+----------

     0     |    5555  |   1000 |   8FFF      1     |   12345  |     20 |   9080

     2     |      FF  |      1 |    200                    (Fig.-1)

This table shows the Last LBA of Segment ID 0 is 5555h and this data (LBA:5555h) is  stored at Page 8fffh on Sector Buffer and the Length of this segment is 1000h. It means  that the segment contains LBA 4556h - 5555h and it is stored at Page 8000h - 8fffh. Above table also shows Segment ID 1 has the data of LBA 12326h - 12345h stored at Page  9061h - 9080h, Segment ID 2 has the data of LBA ffh stored at Page 200h. 

When HDD receives a READ COMMAND such as READ SECTORs, READ MULTIPLE, READ DMA and so on,  HDC searches Segment Informati...