[Tile-serving] Parallelizing more of osm2pgsql

[Kai Krueger]

On 07/19/2013 10:13 AM, Sven Geggus wrote:
> Kai Krueger <kakrueger at gmail.com> wrote:
> 
>> Overall, this results in about 20 - 30% speed up of the complete import
>> time on a standard slim mode import and nearly halves the time needed to
>> import an extract with the --drop option (dependent on hardware).
> 
> Does this speed up incremental updates as well?

I don't know yet, as I haven't worked on incremental updates till now.
In fact I am not even sure if the incremental updates aren't in the
current state of the code is in broken completely.

However, the same paths that I have been parallelizing for the initial
import are also mostly used for the incremental updates, so it is well
possible that it speeds those up as well. At least I hope it will.

On the other hand diff imports are seldomly CPU bound, and depending on
where the bottleneck is, parallelizing things might not help as much.

With the main imports now seeming to work without (known) issues, I am
going to work on ensuring that the diff imports work next. I'll keep you
updated once I figure out what effect the parallelization has on them.

Kai

> 
> Sven
>