You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Map clustering is one of our biggest performance issues right now, because (a) we need to transfer all of the pin data to the client in order to create the clusters, and (b) once the data is on the client, it takes a long time for the browser to create the clusters.
Action Items
Attempt to transfer clustering to the server.
Resources
This article contains a demo showing 6 million points clustered with leaflet using the supercluster library.
It also says the clusters can be created server-side using Node. Unfortunately it doesn't explain how to do that, except that you need to use the vt-pbf package to convert the clusters to Mapbox tiles.
The supercluster repo contains a demo that uses a webworker for clustering. Maybe the code in the webworker, which makes the api calls, could be transfered to the server.
Supercluster is written in javascript, but here's a python port of the package that claims to be super-fast.
And here's a demo for using supercluster without a webworker.
The text was updated successfully, but these errors were encountered:
Description
Map clustering is one of our biggest performance issues right now, because (a) we need to transfer all of the pin data to the client in order to create the clusters, and (b) once the data is on the client, it takes a long time for the browser to create the clusters.
Action Items
Resources
This article contains a demo showing 6 million points clustered with leaflet using the supercluster library.
It also says the clusters can be created server-side using Node. Unfortunately it doesn't explain how to do that, except that you need to use the vt-pbf package to convert the clusters to Mapbox tiles.
The supercluster repo contains a demo that uses a webworker for clustering. Maybe the code in the webworker, which makes the api calls, could be transfered to the server.
Supercluster is written in javascript, but here's a python port of the package that claims to be super-fast.
And here's a demo for using supercluster without a webworker.
The text was updated successfully, but these errors were encountered: