What's new in PDI 4.0
Index
Introduction
PDI 4.0 is a nicely balanced release, a rare mix of a lot of new features combined with engine stability and 100% backward compatibility of your existing jobs and transformations.
Once again, many many thanks go to our large community of Kettle enthusiasts for all the help they provided to make this release another success.
General changes
Visual changes
Mouse-over
More intuitive menus
New welcome screen
Hop creation
Improved error handling configuration
New perspectives support for Agile BI visualisations, modelling, scheduling, etc.
Running jobs in Spoon
Drill down into running job entries
Visual indicators of running and completed job entries with success and failure mini-icons
Mouse over completion mini-icons shows details of execution results
Running transformations in Spoon
Drill down into running transformation job entries and mappings
Row input/output sniff testing: see what rows are passing
Remote input/output sniff testing on a Carte server
New logging architecture
Reduced memory consumption
Incremental log updates
Global log buffer size limit for long running jobs/transformations
Interval logging
Auto clean-up of old log records
Log record time-outs
Log record lineage
Log record colour coding in Spoon (blue and red for error lines)
Step Logging
Job entry logging
Execution lineage logging
Renaming individual columns
Global configuration options for all log tables
New plug-in architecture
Unified plug-in architecture
Easier deployment and packaging
Step, job entry, partitioner, database type, spoon perspective, life-cycle, ... : all pluggable
New repository plug-in architecture
Allowing for 3rd party repositories like the Pentaho Unified Enterprise Repository
Removed dependencies to relational database repository (still supported though)
Added support for repositories capable of team-development (file locking)
Added support for repositories capable of fine-grained security repositories
Added support for repositories capable of storing and retrieving revision history
Step changes
New steps
SAP Input: Reads data from an SAP/R3 application server. (needs jsapco.jar not included in PDI)
Data Grid : Allows you to enter static rows of data for reference or testing purposes
OLAP Input: read data from an OLAP server using olap4j over XML/A: Mondrian, Palo, SSAS, SAP B/W
Salesforce Delete, Insert, Update, Upsert
Add fields changing sequence: a sequence that gets reset when the values in a set of fields changes. (group sequence)
User Defined Java Class: create your own plugin on the fly in a step (coming out of incubation)
Send information using Syslog: Send a message to a Syslog server. http://en.wikipedia.org/wiki/Syslog
Java Filter : Filter based on a User Defined Java Expression
Memory Group By: for smaller groups you can keep the intermediate statistical results in memory leading to faster results
LucidDB streaming bulk loader
Teradata Fastload Bulk loader
Experimental steps added: Get table names, Email messages input, ...