As I train customers, someone will mention that a critical piece of information is stored in a spreadsheet and that is the primary store for it or that users export data from Maximo and manipulate it in Microsoft Excel.

I have seen spreadsheets contain:

• Details of integration messages that have/haven’t been processed
• Maximo/Interface configurations
• Passwords
• Large numbers of records exported from Maximo for analysis that could be performed in Maximo
• Data loading

It depresses me to hear this because of the spreadsheet related problems I have seen.

Spreadsheets have their place but in my experience they can cause a significant number of problems.

Problems that include:

1. Data is manually copied from Maximo into spreadsheets so a couple of extra fields could be populated e.g. user who processed the message  – A better solution is to modify the Maximo so that it stores the extra details
2. Configurations that are out of date – BIRT reports can be built that provide this information from the database and ensure that other administrators always have the most up to date information
3. Passwords – The passwords on Microsoft Excel files have been easy to crack and there will always be people interested in selling software to break the passwords
4. Database performance problems because the database is very busy retrieving the records for the export
5. JVM performance problems because the JVM is busy performing calculations in its memory
6. JVM crash because the JVM does not have enough memory to handle the amount of data / how it is handled

What can you do as a system administrator?

• Understand your user processes and challenge them when they are relying on spreadsheets too heavily
• Look for evidence in the logs of large downloads/queries – Websphere Access.log and Maximo’s SytemOut.log can both show when this is happening
• Consider if extra fields/validations/customisations in Maximo can remove the need for spreadsheets
• Consider if a Cognos or BIRT report can be built that does the same work
• On a non-Maximo system I delivered a solution that reduced 2.5 days of manual work to a scheduled report with 30 minutes follow-up work.
• Consider if Cognos or BIRT reports can be customised to push more of the calculations onto the database so they are not performed in memory
• Understand the best practices around loading data e.g. restarting the JVM after large data loads. I'll talk about dataload considerations in a later blog posting

This blog series

This article is one of a series of articles to help system administrators understand the Maximo logs and the underlying architecture.

This series is written by Mark Robbins who has been recognised by IBM as an expert in his field via the IBM Champion programme. Mark has supported/developed Maximo systems and trained system administrators for over 8 years and actively promotes good support practices via his blog and webexes.

Articles are normally posted on a Tuesday.

If you like this article then please share or like it.

Whilst I support the wider Maximo community and encourage the spread of knowledge, when republishing content from this blog please include the originating author along with the article or parts of.

If people do find parts of this blog coming up in blogs/newsletters/communications then please contact me directly.