[OSM-dev] Slow rendering

[Stephan Knauss]

On 04.05.2021 18:49, Alex Chirigiu wrote:
> On 04-May-21 17:32, Stephan Knauss wrote:
>> I hope that Alex is going to share some figures and optimization 
>> results here on the list.
>> So @Alex: Even with you potentially paying someone for optimizing 
>> your rendering stack, would be great if you could share the results 
>> so others could benefit in the future as well. 
> -------------------
>
> It might also be ESXI related. If I will find something, I will share it.
> That machine acts strange since it won't saturate anything else than RAM.
>
> *3) Nothing at all on the CPU *(this is nailed down at 39.xx% usage)


You mentioned that you have 40 cores available. And the machine has with 
256GB plenty of RAM. potentially enough to keep the full extract of 
Europe in RAM. You have roughly 5 minutes i/o load at 300 MB/s. 90 GB 
might be the size of the DB which is then cached and would explain why 
i/o drops.

Please check on database size.

You started rendering of zoom 15 covering the full globe. So roughly 16 
million metatiles. The majority of these tiles will simply be empty, as 
they are just water or in an area where you have no data.

Europe is roughly 500k metatiles, depending on bounding box. So you 
rendered 32 times more metatiles than needed. This is something you 
could optimize. On higher zooms it will get worse.

You see a CPU load of around 39% with 16 cores. On 40 cores this would 
be nearly 100%. So I guess that you are limited here by single-thread 
performance of the active threads. Could you check process statistics 
here? Which process is at 100%? is it renderd or postgresql?

Stephan