Browse Prior Art Database

A method for handling big meta data on file system

IP.com Disclosure Number: IPCOM000236770D
Publication Date: 2014-May-15
Document File: 1 page(s) / 11K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method for handling big meta data on file system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 86% of the total text.

Page 01 of 1

A method for handling big meta data on file system

A method for handling big meta data on file system

Disclosed is a method for handling big meta data on file system.

There is a method for using a tape medium (a tape cartridge or medium) via a file system as if the tape medium is a HDD or a USB memory. For example, a linear tape file system (LTFS) manages a tape medium split into two areas called a data partition (DP) and an index partition (IP). The LTFS stores the main body of a file in the DP of a tape medium and also stores metadata on the file, called an index, such as a file name or an ID for identifying the file, in the IP of the tape medium in an extensible markup language (XML) file format. The index includes metadata on all of files stored in a single tape medium. The size of the index is about 1 GB per 1 million files. The maximum size of each metadata depends on the type of the metadata. For example, the size of fork such as Alternate Data Stream (ADS) and resource fork may be larger than the size of file body. So it is hard to store index which has fork in the IP. Actually, LTFS does not support fork.

The disclosure provide the method for supporting fork for LTFS by storing different data in index on IP from data in index on DP. Concretely, large size metadata is stored only in index on DP. By not to large metadata in index on IP, supporting fork has no effect for free capacity on IP.

For threshold of large metadata, 4 KB or other fixed value is usa...