[
https://issues.apache.org/jira/browse/LUCENE-7270?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15268797#comment-15268797
]
Karl Wright commented on LUCENE-7270:
-------------------------------------
Implemented this and discovered that actually my former implementation
performed much better for the London borough case. In fact, the 2D balanced
tree is 4x slower:
{code}
ITER 0: 0.05 M hits/sec, 0.61 QPS (53.91 sec for 33 queries), totHits=2693539
{code}
I believe this may be because it is visiting far more nodes (but not more
edges), because of the way the ranging is done:
{code}
private static boolean traverse(final Node node, final EdgeIterator
edgeIterator, final double minValue, final double maxValue) {
if (node == null) {
return true;
}
if (minValue <= node.max) {
// Does this node overlap?
if (minValue <= node.high && maxValue >= node.low) {
if (edgeIterator.matches(node.edge) == false) {
return false;
}
}
if (traverse(node.left, edgeIterator, minValue, maxValue) == false) {
return false;
}
if (traverse(node.right, edgeIterator, minValue, maxValue) == false) {
return false;
}
}
return true;
}
{code}
I'm a bit puzzled as to why this doesn't totally sink the 2D code. Maybe it's
because the first edge it finds causes it to exit? Or maybe I just have a
silly error somewhere? Don't know, maybe [~rcmuir] has an idea? I'll attach a
patch if requested.
> Use better balanced trees for Geo3d complex polygons
> ----------------------------------------------------
>
> Key: LUCENE-7270
> URL: https://issues.apache.org/jira/browse/LUCENE-7270
> Project: Lucene - Core
> Issue Type: Improvement
> Components: modules/spatial3d
> Reporter: Karl Wright
> Assignee: Karl Wright
>
> The current tree data structure in GeoComplexPolygon has a lot of weaknesses.
> A better algorithm maybe can be taken from Polygon2D, which basically does
> the following:
> Each node has:
> - low value (which is for that edge alone)
> - max value (which is for that edge and all children)
> Balanced tree building:
> - sort by low value (which is for the individual edge), and use max value as
> tie breaker (which is max for edge and all children)
> - build tree after sorting, picking midpoint and recursively building lesser
> and greater children that way
> Balanced tree traversal (looking for range minValue -> maxValue):
> - Descend the entire tree until the node fails this test:
> if (minValue <= max) { ... }
> So if the minimum value being sought is greater than the max for this edge
> and all children, we stop and don't look at children.
> (Q: does this represent a good split and a targeted range? Maybe... We
> can certainly try it.)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
