> Running in single-node-mode with an _in memory_ database (for our integration tests) is still slower than postgresql. Can you share your numbers to have some

> Running in single-node-mode with an _in memory_ database (for our integration tests) is still slower than postgresql. Can you share your numbers to have some kind of order of magnitude for this difference?