 | Level: Intermediate Mark Baldridge , Principal Consultant, North American Lab Services, IBM
30 Nov 2006 In this installment of the UniVerse®
performance series, continue the investigation of UniVerse file performance,
examining the inner workings of dynamic linear hash files. Explore the challenges
of setting an appropriate separation, once you choose one.
Objectives - Learn the mechanism of dynamic file splitting and the
modified definition of an oversize record in a dynamic file.
- Measure
the split operation to compare it to overflow in Part
2 of this series.
- Learn the impact of loading an empty dynamic file and MINIMUM.MODULUS.
- Learn the limitations of GROUP.SIZE and how to overcome them when setting a separation.
Prerequisites
It is helpful to have completed the previous parts of this tutorial series,
but not required.
System requirements
This tutorial requires an installation of UniVerse. You can download a Personal
Edition of UniVerse for Red Hat Linux® or Microsoft Windows®
from the IBM Web site. The personal edition has file size
restrictions that will prohibit some tests, such as test 1332, which creates a 1GB file. You should have a familiarity with writing and compiling UniVerse BASIC programs.
Formats html, pdf
About this series
This series
of tutorials provides UniVerse DBA and application designers with tools for making reasoned and metrically-justified decisions about choices in everyday tasks. The subject matter of this series of articles and tutorials on UniVerse performance tuning arise from visits to customer sites. The issues raised do not disparage the customer, but reflect a real-world predicament. The typical developer or DBA has so many outstanding tasks that by the time one task is complete, at least one more has joined those remaining. They all demand attention. Sufficient time exists to make something work, but typically not enough to make it work well.
Tutorial overview
With this installment of the tutorial series, continue to explore the factors that affect file
performance. This tutorial starts by examining the file structure of dynamic files
so you can see how they relate to the structure of static hash files. It then
explores the dynamic file-splitting operation so you can make measurements of the
process. And it wraps up with techniques that overcome the default separation and limits placed upon that selection for dynamic files.
| | | |
