![]() ![]() ![]() The main reason is that we spent much time optimizing some of the critical AQL code paths. There are several reasons why the 2.6 performance is better than in previous versions. The conclusion is thus that the speedups did not depend much on the dataset sizes. In some cases, the speedups varied a bit with the dataset sizes, but it was still in the same ballpark for all three datasets. Speedups were observed for all three test dataset sizes alike. Still 18 to 25 % seem okay as a free take-away. The speedups observed for this type of query are between 18 and 25 %, which is the lowest speedup measured by the tests. The one thing that did not change much when comparing 2.6 with 2.5 is iterating over a collection and returning all its documents unmodified.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |