Re:Jepsen: Radix DLT 1.0-Beta.35.1
by zbentley on Monday February 07, 2022 @12:58AM
(#30240967)
> Isn't this is a rather common configuration for a distributed DB?
In my experience it is not. Rather, aggressive fsyncing/O_DIRECT usage are common. The rationale for this is usually partition risk: better to durably log a write before propagating it than to potentially fail in propagating it and then be left in a position of having to either reactively fsync or hope that automatic flush-to-disk will persist your unexpectedly-sole possession of that update.
In my experience it is not. Rather, aggressive fsyncing/O_DIRECT usage are common. The rationale for this is usually partition risk: better to durably log a write before propagating it than to potentially fail in propagating it and then be left in a position of having to either reactively fsync or hope that automatic flush-to-disk will persist your unexpectedly-sole possession of that update.