Abstract

This article describes how to “discover” IBM Informix instances on a specified host and then gather further information about databases within each discovered instance.

[Read More…]

Abstract

One major factor to obtaining and sustaining good performance within an IBM Informix database is having the appropriate statistics available. This article provides another method and explores the “output to pipe” utility within SQL. There are several ways that statistics can be maintained, the most common, apart from home grown, are:

• AUS – Auto Update Statistics, driven by the dbscheduler and configured most easily through OAT
  • http://pic.dhe.ibm.com/infocenter/informix/v121/index.jsp?topic=%2Fcom.ibm.perf.doc%2Fids_prf_742.htm
• Art Kagel’s dostats package – widely used in the IBM Informix Community and comprehensive
  • https://sourceforge.net/projects/iiug-software-repository/files/DBAdmin-Tools/utils2_ak

What is focused on here is having a simple method to generate and run efficient update statistics commands providing the recommended set of required statistics without getting into setting up AUS or compilation of dostats – i.e. a simple way to ensure your database has the relevant statistics generated using a single command; ideal for test environments or small production environments. This also makes use of the IBM Informix SQL statement “output to pipe” which is used to pass output to another program, in this example, “awk”.

[Read More…]

Abstract

Avoiding sub-queries can be essential to reduce long run-times on large data sets. However, it is sometimes necessary to identify rows in one table for which there is no associated row in another table. An example would be an order for which there are no order detail lines. The conventional approach is to use the NOT EXISTS (sub-query), but this can be extremely slow for large data sets. This article demonstrates how an ANSI join can achieve the same result much more efficiently, even on large data sets.

[Read More…]