Bulk loading (with STR) about rtree HOT 6 CLOSED

hi @davidmoten

Can you review my implementation of this feature?

from rtree.

Thanks I'll have a look. If the data is static you can use serialisation by
the way. Look at README.md.

On Thu, 14 Apr 2016 01:27 x2b [email protected] wrote:

I have a rather large set of entries (>= 300k) and I would like to
somewhat improve the performance of the tree. The data is available
beforehand so I think that bulk loading might help, in particular the
"Sort-Tile-Recursive" strategy. I think this could be implemented fairly
easily.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub
#45

from rtree.

Do you observe a speed improvement relative to the RTree creation?
I also need to handle very large datasets and the RTree creation time is my biggest issue.

Update: sorry forget my comment I have just seen that the comparison was made in the related fork:
#67

Thanks!

from rtree.

Hi, actually I am still working on the performance evaluation. Sorry for the delay.

I have tried to add more test cases in this benchmark class in this branch, however it seems to mess up the codebase and it is very tedious to make some changes to the rtree types or datasets.

So I opened a new project according to the guide of jmh, and trying to make the code more clean with help of the inheritance feature.

I am no familiar with jmh, so I would greatly appreciate if you guys have any suggestions on this benchmark project.

BTW, maybe this issue is safe to close and leave me some more days to open a new PR to update the README docs after I have mature benchmarking results?

from rtree.

ok closing, thanks for your efforts @ambling

from rtree.

@ambling: Thank you for setting up this benchmarking project.
I'll try to have a look at it.

from rtree.