26.02.2020 17:35, Paul Reeves wrote:
The idea of the failsafe comparison is valid but it depends on how they have configured journalling.
The common trick for benchmarking is to set up a very long (practically infinite from the test POV) checkpoint interval. It provides the best possible performance (at the cost of the worst recovery time, but who cares).
Dmitry Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
