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

Time Based Compression Selection Line Detector Methodology

IP.com Disclosure Number: IPCOM000110394D
Original Publication Date: 1992-Nov-01
Included in the Prior Art Database: 2005-Mar-25
Document File: 1 page(s) / 48K

Publishing Venue

IBM

Related People

Duvall, KE: AUTHOR [+6]

Abstract

Line load variations in a distributed environment cause significant performance variations over time. Further, transient load conditions produce conditions of saturation and unacceptable performance at stochastic (unpredictable) times.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 77% of the total text.

Time Based Compression Selection Line Detector Methodology

       Line load variations in a distributed environment cause
significant performance variations over time.  Further, transient
load conditions produce conditions of saturation and unacceptable
performance at stochastic (unpredictable) times.

      In the AIX* environment, a polling technique to determine
current LAN and TTY conditions at periodic times was created.  The
basic application methodology is best described as follows:
      A test buffer is timed for a particular selected production
path to determine the current duration of a selected data block from
a candidate file object from a managed storage group.
      The client workstation in an ESMS environment polls the LAN and
host connections periodically (default every 10 minutes).  The
polling methodology captures a fixed buffer transfer time, and
compares this time to the time required to compress the same buffer.
      The algorithm projects the time required to compress a given
data type as a function of the current utilization of a selected
path.  Alternate paths are selected and polled if available in the
configuration.
      A value summary is maintained by the ESMS mover method in each
client for current transfer times.  In this way, a decision can
automatically be made regarding compression function execution on the
workstation.  If the workstation connection is optimal, then minimal
time and resource will be co...