Browse Prior Art Database

Multi-Level DASD Array Storage System

IP.com Disclosure Number: IPCOM000115628D
Original Publication Date: 1995-Jun-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 2 page(s) / 69K

Publishing Venue

IBM

Related People

Ouchi, NK: AUTHOR

Abstract

Disclosed is a multi-level DASD array where the array domain characteristics are optimized to the performance/cost requirements of the level of the storage system.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Multi-Level DASD Array Storage System

      Disclosed is a multi-level DASD array where the array domain
characteristics are optimized to the performance/cost requirements of
the level of the storage system.

      DASD array characteristics such as domain width, block size,
RAID level, device capacity, and other factors such as data
compaction affect the performance and cost of the net storage system
to the end user.  Storage hierarchies are an architecture that
permits the storage system to be a composite of several storage
technologies to optimize the performance and cost of the total
storage system.  The use of a semiconductor memory as a cache for
DASD is one example of a storage hierarchy where the high performance
(and cost) of a small amount of semiconductor storage can
significantly increase the overall performance of the DASD subsystem
while the overall cost of the system is predominantly the cost of the
DASD devices.  Disclosed is a multi-level DASD array where the array
characteristics are different for levels of the storage system to
emphasize either performance or cost so that the composite system has
the performance characteristics much like the performance level with
cost much like the cost level.

      DASD performance can be affected by many factors.  One may be
block size.  Small blocks permit rapid access to random data but
consume capacity because of the larger number of inter-record.
Domain size, RAID level, etc. as well as data compaction, trade
performance and cost.  Fig. 1 il...