I'm new to leafletjs. Been working on cesiumjs for a while and we are trying leaflet now. The main reason for the switch is to see if there's a huge performance difference.
In Cesium, I had a collection of primitive points that I plotted. What's the most efficient way of plotting 140K points in leafletjs? Using markers or creating individual little circles?
I am also thinking of using the clustering plugin (http://leafletjs.com/2012/08/20/guest-post-markerclusterer-0-1-released.html), so please share any thoughts on performance.
You have 2 common options:
Display your points in a Canvas-based layer, like using Circle Markers and force rendering them on a Canvas instead of SVG (see also Leaflet MaskCanvas plugin). Circle Markers, even on a canvas, still emit events, so you can detect "click" etc.
Use a clustering plugin, like Leaflet.markercluster plugin that you mention for exampe. It can handle your 140k points, depending on the client's computer performance (see https://github.com/Leaflet/Leaflet.markercluster#handling-lots-of-markers and demo http://leaflet.github.io/Leaflet.markercluster/example/marker-clustering-realworld.50000.html with 50k points, but note that the demo uses an old version of the plugin, whereas the current version is even faster).
Trying to display your 140k points without Canvas or clustering will crash your browser for sure.
If you want to render more than 100k markers, you can use Supercluster library, because Leaflet.markercluster loading of >100k markers could take more than 30 seconds.
I created a github repo to compare initial loading of Leaflet.markercluster and Supercluster.
Related
I'm trying to add real-time, web-socket event-based data to my map.
Everytime I get a point, I must add it/update it on map.
What's the best option?
A) Create a FeatureCollection and add a source and respective layer. When updating, change the FeatureCollection and call setData();
B) For every point, create a different source and layer. When updating, just change the respective source and call setData();
I really don't believe B) is the best option, I'm just not sure about the perfomance of option A) (or I'm thinking about the former the wrong way).
I'd say it somewhere between the two. For Mapbox Draw I use two layers, one for features that are being edited and one for features that aren't changing. In your case, you'd want to create a layer for every 100 features. This is because Mapbox GL has to recut the geojson into tiles every time you add a feature so limiting the number of features that have to be reviewed is wise. That said, lots of layers will be a problem too.
While I said 100 features above, you'll want to play around with this number a bunch. It will be about finding the right balance between number of layers and number of features in a source.
So, basic gist is, I have my own tiles of not the real world I'd like to display with the Google Maps viewer. I've found examples of how to split an existing single image into tiles for use with it, but nothing that deals with setting up your own tiler.
I have map data such as this:
https://dl.dropboxusercontent.com/u/44766482/superimage/index.html
Which right now is just a bunch of 1600x1600 images in an html table. This naive method works, but I'd like to switch to the more robust google api for better zooming and smarter streaming of the image data.
I've been unable to find a good example of how to generate your own tiles for the zoom levels and bring it together with some html/js.
If you want some more information for the goal process;
I have a python script that can output any size tiles of the map, at any zoom level. I'd like to bundle those together into a google maps api website. But in my own efforts I have not found a good example or documentation of how to do that. For one, I can only find examples of how the zoom levels work for the real world map, but not for a custom one.
Edit:
Got most things working as I want them, but I'm still confused regarding the "center" that can be set, as it's in lat and lng, which don't apply. I'd also like to set boundaries as currently it tries to load .png files outside of the maps range.
My current progress:
https://dl.dropboxusercontent.com/u/44766482/googlemapspreview/index.html
I think what you are looking for is the google maps imageMapTypes:
https://developers.google.com/maps/documentation/javascript/maptypes#ImageMapTypes
Basically, each zoom level is the 4 lower zoom tiles combined. A Projection function can be skipped to get orthogonal mapping.
I'm looking at plotting a scatterplot with a large number of points (500,000 and upwards).
Currently, we're doing this in Python with Matplotlib. It plots the points, and it provides controls to pan and zoom. I don't believe it provides any clustering or points, it just plots them all - doesn't make much sense at the zoomed out view, I suppose, but you can zoom in and they're all there.
I was looking at doing the chart in JavaScript, to make it a bit easier to distribute. I was looking at D3.js, to see if something similar is feasible there. I did find this example of a basic scatterplot:
http://bl.ocks.org/mbostock/3887118
Firstly, would you be able to plot that number of points? (500,000 and upwards) I was under the impression you couldn't due to the overhead of all the DOM objects? Are there ways around this?
Secondly, is there any kind of clustering available, either a library or even just an example of this being done in D3.js?
Thirdly, if anybody knows any good examples of pan/zoom functionality and clustering, or even just a packaged JS library that handles it, that would be awesome.
Fourth, it would be also nice to have click handlers for each point - and to display some text either in a overlay, or even just in a separate window. Any thoughts on this?
Can you draw half a million points with D3? Sure, but not with SVG. You'll have to use canvas (here's a simple example with 10,000 points that includes brush-based selection: http://bl.ocks.org/emeeks/306e64e0d687a4374bcd) and that means that you no longer have individual elements to assign click handlers to. You will not be able to render half a million points with SVG, because all those DOM elements will choke your interface, as you mentioned.
D3 does include quadtree support that can be leveraged for clustering. It's in use in the above example to speed up search but you could use it to nest elements in proximity at certain scales.
Ultimately, your choices are:
1) Some other library/custom implementation that renders in canvas and polls the mouse position to give you the data element rendered at that point.
2) A sophisticated custom D3 approach that nests elements in proximity and only renders SVG elements appropriate at the zoom level and canvas position (pan) you're at.
Yes, D3.js can be made to work with million scale data with two things:
pre-rendering on the server side. For more see here: https://mango-is.com/blog/engineering/pre-render-d3-js-charts-at-server-side/
By aggregating (or clustering) part of the data so that user can interact and expand the graph if need be. For this use collapsible nodes if you can (http://bl.ocks.org/mbostock/1062288).
Also avoid using force layout. It takes time to settle and converge to a stable positioning.
For clustering libraries, I would pick one up off the shelf. I would choose the scikits library from python, there are many in JavaScript but they are not very robust as they mostly cover k-means or hierarchical clustering. I would precalculate the coordinates using scikits by clustering and then render it using D3.
D3 handles Pan and zoom. Again click handlers and text display are available in D3. (http://bl.ocks.org/robschmuecker/7880033)
I'm showing the boundaries of cyprus regions using openlayers with vector layer, I get the coords from a shapefile which has a lot of points, for instance some regions can have a multipolygon with in total :30000 Points.
This is really slow on firefox / IE and in firefox it shows unresponsive script error. Only chrome works well. What's the best way to get it working faster and smoothly ?
I read one answer is to create a WMS server with MapServer to generate the images tiles instead to use the OpenLayers directly but I can't implement that solution because I want to change layer style dynamically to the different features in the vector layer i.e to show different colors to the region based on the dynamic value.
How can I overcome this issue. Any Insights would be much appreciated.
I think WMS + TileCache will helps to improve the speed. You mentioned you want to change layer style dynamically, I think there are also some solution to do that with WMS, please check:
https://gis.stackexchange.com/questions/64113/how-to-dynamically-change-sld-style-of-wms-layer-being-served-by-geoserver-from
and http://openlayers.org/dev/examples/sld.html
I'd say that GeoJSON may increase your performance in Firefox. WMS is the best in my opinion, and if you work with mapserver you wouldn't have problems to do what you want, geoserver may help you also, but for dynamic changes I'd say you'd need GeoExt.
Hope this helps,
Just to present a contrarian view, sending this as a vector layer but thinning the data points at distant zoom levels would provide the performance benefits you need with the interactivity you want.
We started using Google Maps on our web application rather extensively. It worked fine at the beginning, but as we add more markers we find that the performance are not quite there. Although I'm quite sure we don't use it in the most efficient way.
I am looking for information about Google Maps best practices and tips'n tricks. Any suggestions?
You might find some good ideas in this article, which compares several methods of handling large amounts of markers.
Marker Manager has some limitations, depending on what you're trying to accomplish; for instance, it doesn't allow every marker to be available from every zoom level. I created a clustering function based on the principles discussed in this tutorial. It uses the Static Maps API in PHP, but the principles behind the clustering can be used however you want.
Update: This clustering utility was just released: MarkerClusterer
Use Marker Manager.
Limit markers to what's visible (ie, understand the window boundaries, and only show markers that fall inside the window)
Learn to listen for various map activities and react - such as viewpoint moves, zooming, etc - to update the markers
Don't show markers that overlap significantly - show only one marker (perhaps a different shade or color to denote there are several points at this marker) and let the user zoom in if they want to see the individual markers. Use the tooltip to show a zoomed in window if you want to get fancy.