Benchmark Factory for Databases is a database performance and code scalability testing tool that simulates users executing transactions on the database and replays production workloads in non-production environments.
What’s new in this version
This release of Benchmark Factory includes the following new features and enhancements.
Industry Standard Benchmark Testing
Resume Create Objects Step After Interruption
* If there is an interruption in the process of loading data during the Create Objects step (for example, the database runs out of space or the connection is lost), you can now resume the process without losing the objects and data created thus far.
This is useful as it allows you to resolve the issue that may have caused the interruption (for example, by allocating more space or re-establishing the connection) and then resume the job.
* This enhancement also allows you to manually stop and resume a job during the Create Objects step without losing the objects and data created thus far in the test.
Increase Scale Factor After Interruption (Create Objects Step)
* If there is an interruption in the Create Objects step, Benchmark Factory allows you to increase the scale factor before resuming the process. Please note that this feature does not support a decrease in the scale factor.
Note: If you decrease the scale factor, Benchmark Factory cannot resume the process and must reload the dataset using the decreased scale factor.
Improved Performance When Loading Data
* This release includes several enhancements that reduce the time it takes to load benchmark objects, especially when using a large scale factor.
When you run an Industry Standard Benchmark test, Benchmark Factory creates database objects (tables/indexes, etc.) and loads them with data. The number of objects created and the amount of data loaded is determined, in part, by the size of the scale factor you specify. Large scale factors require large amounts of data for testing and can lead to long data loading times. This enhancement helps reduce the time necessary to load benchmark objects when a large scale factor is used.