Browse Prior Art Database

Method for Translating Logical Block Addresses to Physical Addresses on Magneto-optic Media

IP.com Disclosure Number: IPCOM000109496D
Original Publication Date: 1992-Sep-01
Included in the Prior Art Database: 2005-Mar-24
Document File: 6 page(s) / 244K

Publishing Venue

IBM

Related People

Bish, JE: AUTHOR [+3]

Abstract

On an Optical Media drive the attached Host will request access to data that resides on the media. The request consists of a Command, a Logical Block Address and a transfer length. The Host views the media as a contiguous series of data sectors with are addressed by Logical Block Addresses. However, the media consists of sectors which are addressed by a physical address consisting of a track value and a sector offset from the start of the track. The total data space on the media is divided into groups which consist of data sectors and spare sectors. The spare sectors are used in the event that a data sector is found to be defective.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 36% of the total text.

Method for Translating Logical Block Addresses to Physical Addresses on Magneto-optic Media

       On an Optical Media drive the attached Host will request
access to data that resides on the media.  The request consists of a
Command, a Logical Block Address and a transfer length.  The Host
views the media as a contiguous series of data sectors with are
addressed by Logical Block Addresses.  However, the media consists of
sectors which are addressed by a physical address consisting of a
track value and a sector offset from the start of the track.  The
total data space on the media is divided into groups which consist of
data sectors and spare sectors.  The spare sectors are used in the
event that a data sector is found to be defective.  This article
describes a method for translating a Logical Block Address into a
physical address which is used to access the proper data and a method
for converting the transfer length into smaller physical segments (if
necessary) to avoid defective sectors and group boundaries.

      Fig. 1 shows a typical layout of the track and sector structure
of the media discussed here.  Each row in the figure represents a
track on 1024 byte sector media.  Each square represents a sector.
There are 17 (0 through 16) sectors per track.  Surface analysis has
been performed on this representation.  Primary defects are not
counted during group assignments, thereby causing the group
boundaries to be moved out or slipped by the number of primary
defects.  Also shown, are defective sectors, both primary and
secondary, which have been tabulated in the defect tables (see Figs.
2 and 3).

      The Defect Management Tables are compiled and stored on the
media according to a media standard.

      When new media is inserted into the drive or the drive is
powered on with media installed, the initialization process will read
the Defect Tables into RAM.

      The Secondary Defect List is copied to the media every time a
new defect is found.  Therefore, the Secondary Defect List will be
maintained in RAM for the duration that the media remains in the
drive.

      The Primary Defect List is written once to the media during
initialization and certification and never written again.  The
Primary Defect List is placed into work RAM whenever the medium is
mounted and discarded whenever the medium is removed.

      There are two components to a host request when accessing data.
The first component is the Requested Logical Block Address (RLBA) and
the second is the Requested Transfer Length (number of blocks).  For
example, the host may request that 40 blocks be read beginning at LBA
32.

      When a request for such a data access is received, the Defect
Manager must be able to locate the physical address to which LBA 32
is mapped to and (if necessary) break down the 40 block request into
smaller physical segments which filter out all known defects (those
tracked in the Primary Defect List or...