Memory Optimizations (Revision #54)

[All tests were done with Windows 32bit versions of Aqua Data Studio]

Version 8.0 of Aqua Data Studio now includes improved memory optimization methods than previous versions. One of these techniques is a memory compression method for optimizing resultsets which includes redundant data such as numeric data. Quering for large results in data warehousing environments with large quantities of numeric data will yield dramatic memory reductions. Long term use of Aqua Data Studio in these environments will keep memory usage low and in turn keep performance of the application high. The optimizations are most apparent in the Query Analyzer, Query Builder, Saving Results and Exporting of data.

Below are simple examples of a single query on two different types of result sets. One is of redundant numeric and text data, and the other is non-redundant data. The memory optimizations depend on the datasets, but these examples show some relative numbers which can be used to compare optimizations on different datasets.

Memory Consumption on Large Results Sets using the Grid Results

Version 8.0 uses les memory than its predecessors on queries with large resultsets of "redundant" data in the Grid results. Older versions required 220% | 455% more memory.


v6.5	v7.0	v7.5	v8.0	ADS Memory Usage	OS Memory Usage by JVM
189MB / 254MB	182MB / 254MB	109MB / 187MB	34MB / 49MB

Version 8.0 uses les memory than its predecessors on queries with large resultsets of "non-redundant" data in the Grid results. Older versions required 12% | 64% more memory.


v6.5	v7.0	v7.5	v8.0	ADS Memory Usage	OS Memory Usage by JVM
173MB / 254MB	187MB / 254MB	118MB / 209MB	105MB / 141MB

Memory Consumption on Large Results Sets using the Text and Grid Results

Version 8.0 uses les memory than its predecessors on queries with large resultsets of "redundant" data in the Text and Grid results. Older versions required 24% | 97% more memory.


v6.5	v7.0	v7.5	v8.0	ADS Memory Usage	OS Memory Usage by JVM
140MB / 159MB	126MB / 224MB	88MB / 145MB	71MB / 77MB

Version 8.0 uses les memory than its predecessors on queries with large resultsets of "non-redundant" data in the Text and Grid results. Older versions required 17% | 20% more memory.