On 5/14/18, Warren Young <war...@etr-usa.com> wrote:
> Your benchmark doesn’t address the primary problems pointed out in the
> Mozilla article:
The point of my article is that many people assume *without measuring*
that reading and writing directly to the filesystem will be faster
than using a database. That assumption is incorrect.
The Mozilla article does not state this assumption explicitly, but it
is there just beneath the surface.
There are indeed many issue to consider when designing a storage
system for an application. My point is to show that a relational
database is often a much better choice than many people (including the
authors of the Mozilla article) believe. Prejudices against
relational databases that were acquired due to bad experiences in the
previous millennium are still quite pervasive. I aim to combat those
prejudices with actual measurements.
Where there any actual measurements in the Mozilla article? It is
quite old and I haven't read it in years, but I don't recall there
D. Richard Hipp
sqlite-users mailing list