Dynamic magnetic information storage or retrieval – General recording or reproducing
Reexamination Certificate
1999-07-30
2001-09-11
Neal, Regina Y. (Department: 2651)
Dynamic magnetic information storage or retrieval
General recording or reproducing
C360S048000, C360S053000
Reexamination Certificate
active
06288862
ABSTRACT:
TECHNICAL FIELD
The present invention relates to methods of distinguishing new information from old information in a medium, and mechanisms that support the methods.
BACKGROUND ART
Mass storage media commonly stores information in sequential recording blocks starting from a loading or initial position and proceeding always in one logical direction. A recording block is the smallest unit that can be written and retrieved from the medium. Recording blocks include, but are not limited to control blocks, data blocks and device blocks. A well known example of mass storage is a magnetic tape used in a streaming tape backup application where the entire image of a hard drive is copied onto the tape. The tape is usually loaded starting at one physical end of the tape. Information is then written in the tape in sequential recording blocks as the tape streams toward its far physical end. In some situations an individual tape may be written on once and then archived indefinitely. Tapes in other situations are reused periodically with the new information being written directly over the old information.
It is well known in the art that the storing of additional information along with the customer's information can be helpful in managing the use of tapes and other media. For example, U.S. Pat. No. 5,485,321 issued to Leonhardt et al., on Jan. 16, 1996 discloses identifiers for the user (customer), system and subsystems, application specifications, physical motion data, and header chronology information stored in a user header accompanying a data record. U.S. Pat. 5,619,384 issued to Leonhardt et al., on Apr. 8, 1997 discloses additional header information including block identification (ID), recording block recovery means indicators, recording block length, sequence numbers within recording block groups and block group ID.
Data about the history of the information stored in the recording blocks is traditionally maintained at the file or volume level where it is applied to large groups of recording blocks, instead of at the individual recording block level in each header. File and volume time information usually identifies when the file/volume was created and when it was modified last. Revision information is often maintained within the file/volume, or is embedded in a file/volume name. In any case, the history is applied to all recording blocks equally. There is no mechanism to identify when the history of any one recording block is out of sync with the history of the surrounding files. This creates problems when reading the information from the medium because old information accidently left in the medium may be output during a read process.
DISCLOSURE OF INVENTION
The invention is a method and mechanism for distinguishing new recording blocks from old recording blocks that are adjacent or intermixed in a medium. The method uses a sequence counter or time stamp, called a Write Pass Count, associated with each recording block within a data volume (a medium and housing) to provide a unique identifier for each occurrence that new information is written into the medium. The Write Pass Count is advanced prior to the start of each write session. During the write session, the Write Pass Count is copied into the header, trailer or some other area where it is associated with each recording block receiving the new information. A Master Write Pass Count for the data volume is maintained on the medium, or in a nonvolatile memory mounted within the housing, if any. (The region in the medium that stores the Master Write Pass Count is called a media information region. The media information region contains data concerning the medium such as media type, history, defect location, indexing, administrative, and other useful information. It is usually found at the logical start of the medium. The media information region is maintained independent of the customer information and is not erased when the customer information is erased.) The Master Write Pass Count is set to an initial value when the data volume is first fabricated and it is never reset. An advantage of this invention is that the Master Write Pass Count and individual Write Pass Counts can be added to existing data volumes without changing the recording format or adding a new physical mechanism to the data volumes.
Write Pass Counts read from the medium can be used in several ways to aid in distinguishing new recording blocks from old recording blocks. In situations where two or more recording blocks have the same recording block identification, the Write Pass Counts associated with the recording blocks can be used to determine the order in which the recording blocks were written. The newest recording block will have the highest Write Pass Count, the oldest recording block will have the lowest Write Pass Count. Where all of the new recording blocks have been written in one write session, all of the new recording blocks are easily identified because they all have the same Write Pass Count. Where the new recording blocks were stored in the medium in multiple write sessions, the Write Pass Counts of the recording blocks will always advance as the recording blocks are read from the first to the last write session of the multiple write sessions. Any recording block having a Write Pass Count that regresses may be rejected as an old recording block that was not properly erased, or is beyond the logical erasure zone.
Accordingly, it is an object of the present invention to provide a method of distinguishing new recording blocks from old recording blocks stored in a medium where the new recording blocks and old recording blocks have the same recording block identifications.
Another object of the present invention is to provide a method for distinguishing new recording blocks from old recording blocks stored in a medium where the new recording blocks are written over the old recording blocks but do not completely erase all of the old recording blocks.
Yet another object of the present invention is to provide a mechanism within a data volume to support the methods of distinguishing the new recording blocks from the old recording blocks.
These and other objects, features and advantages will be readily apparent upon consideration of the following detailed description in conjunction with the accompanying drawings.
REFERENCES:
patent: 4559568 (1985-12-01), Watanabe et al.
patent: 5363264 (1994-11-01), Cavanaugh et al.
patent: 5452284 (1995-09-01), Miyagawa et al.
patent: 5485321 (1996-01-01), Leonhardt et al.
patent: 5619384 (1997-04-01), Leonhardt et al.
patent: 5719717 (1998-02-01), Leonhardt et al.
patent: 5774287 (1998-06-01), Leonhardt et al.
patent: 6031671 (2000-02-01), Ayres
patent: 0 558 169 (1993-09-01), None
patent: 0 936 618 (1999-08-01), None
patent: 60 122426 (1985-06-01), None
patent: 97 27545 (1997-07-01), None
Baron Benjamin Joseph
Gill Richard A.
Wolf James
Brooks & Kushman P.C.
Neal Regina Y.
Storage Technology Corporation
LandOfFree
Method and mechanism to distinguish valid from outdated... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and mechanism to distinguish valid from outdated..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and mechanism to distinguish valid from outdated... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2536586