Browse Prior Art Database

Distributed Disk-Based Queuing and Recovery Mechanism

IP.com Disclosure Number: IPCOM000105817D
Original Publication Date: 1993-Sep-01
Included in the Prior Art Database: 2005-Mar-20
Document File: 2 page(s) / 44K

Publishing Venue

IBM

Related People

Duvall, KE: AUTHOR [+4]

Abstract

Memory intensive queuing on a per-node basis for direct access to a filename or to identifier structures can be reduced by utilizing a disk based queue based upon a multi-level structure definition.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 67% of the total text.

Distributed Disk-Based Queuing and Recovery Mechanism

      Memory intensive queuing on a per-node basis for direct access
to a filename or to identifier structures can be reduced by utilizing
a disk based queue based upon a multi-level structure definition.

      UNIX* directory structures may be used to access a disked based
queue which is directly accessible using a primary request identifier
structure.

      In a UNIX-based distributed environment file server system, a
three-part structure is defined as:

     NodeID   - the identifier of the orginating node for an
operation
     MajorID  - the identifier representing the end-user command
     MinorID  - the identifier(s) representing each of the
subcommands
                required to perform the end-user tasks.

      Using the aforementioned three-part structure, each operation
is uniquely identified.  The queuing mechanism utilizes the base Unix
directory structure in a manner such that each request in the queue
is stored in a flat file, with the name of the file based
(deterministically tied) to the request identifier.

      Access of data associated with a request node can be done
directly via standard system command interfaces to the underlying
UNIX file system, thus taking full advantage of the underlying
caching and hash table lookup operations in the kernel and device
driver logic (thus taking full advantage of 'machine
technology'-optimized search routines...