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

Dynamic archiving algorithm and tool

IP.com Disclosure Number: IPCOM000124725D
Original Publication Date: 2005-May-04
Included in the Prior Art Database: 2005-May-04
Document File: 1 page(s) / 22K

Publishing Venue

IBM

Abstract

In today's distributed computing environment, there are many occasions where files or groups of files need to be moved around between systems. Example: deployment of software from a development system or sandbox to a production or test environment. A number of full-scale (and expensive) solutions exist for maintaining file currency and versioning across systems, but in the absence of such tools, we tend to use the always-installed (and free) standbys like file transfer protocol (FTP). When transferring multiple images, especially across multiple directories, a common practise is to use an archiving tool such as 'tar' to accumulate the files needed, and then FTP or a distributed filesystem implementation to transfer the archive. When frequent updates are being made and tested, it can become tedious and error-prone to reinvoke the same archiving command repeatedly before transferring the archive. This invention proposes a tool which will require the archiving step to be performed only once.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 1

Dynamic archiving algorithm and tool

Proposed is an operating system command with filesystem-level hooks for creating a dynamic archive image. The archive needs to be created (or more precisely, 'defined') only once; thereafter, it can be transferred at will, and will always contain the latest version of its contents from the host system each time it is transferred.

The invention consists of the following:

o An operating system command ("Command") which defines an archive image. The Command uses an algorithm similar to the following: - A list of files to be archived is provided to the Command via syntax similar to the existing Unix 'tar' or 'backup' commands. - The Command creates an archive ("Archive") containing metadata describing those files. The metadata includes information such as the absolute path to each file and the relative path as supplied by the input list (so that restoration can be relative, as with 'tar'). - Assuming the Archive is created on (logically) local storage, the Archive itself will only consume space on disk corresponding to the metadata - it will NOT store the files themselves

o Enhancements to the native filesystem to make it aware of the file type created by the Command, as follows: - The Archive appears in directory listings, stat() results, etc. as an ordinary file.

- The Archive's size in listings, stat() results, etc. will appear to be the sum of [the sizes of the files pointed to by the metadata] and [the metadata itself]. - Whe...