Thanks for the reply, Barry. >From what you’re saying, Fusion Tables don’t seem like a feasible solution >yet. Especially, if the user will have to wait for their View Table to get >created each time they do a search.
I’ve had two major problems – Long delays after map loads until markers get plotted, and poor map performance after markers have been plotted (even with client side clustering). So, from my dataset I’m really just looking the best method to quickly grab the required data from the database and to have them plotted quickly. At most, around 10k markers will be within the view port, so to have high performance when panning/zooming. I saw this example using the Flash Maps api – in terms of map performance with lots of markers it seemed really good...but the Flash Maps api is now deprecated http://www.scribblemaps.com/markerComplex/ How would you tackle this? Thanks From: Barry Hunter Sent: Tuesday, March 06, 2012 12:57 PM To: google-maps-js-api-v3@googlegroups.com Subject: Re: [Google Maps API v3] Re: Performance scaling google maps application On Tue, Mar 6, 2012 at 12:47 PM, Henri Cammiade <henri.cammi...@gmail.com> wrote: Hi Barry, Thanks for the reply - that's what I was trying to understand about the View tables using the Fusion api. I presume each time a View table gets created, it simply adds another table to your account quota (up to five tables)? The "5" - is now many tables you can display at once on a map. A FT account can contain many many tables. I dont know the formal limit, but suspect 1000 So just displaying 1 table at a time - can have many. The quota to worry about is the 250Mb per account. Each View will consume quota - as it in effect takes a copy of the data. (its a materialized view, from traditional database theory) I have a database with over 1 million data points. 1 million markers would NEVER be needed to be mapped at one time. I'm seeking for a fast, high performance solution to a 'dynamic' display as you previously mentioned, where the user selects a location that the map centres and zooms on, then to grab the relevant results from the 1 million to then plot the markers. At most, only around 10k markers would ever be needed to be mapped. This will work. You either need to 1. Build a solution for checking when the table is ready. So the user selects the point - and then has to wait while their personal Fusion Table is built. Once the user has finished, need to arrange to the table to be deleted. or 2. Build 'canned' tables. A series of tables - predefined ready for use. So when the user selects a point, you just pick the approiate table. Maybe use a grid system? Need to keep track of the tables of course! I still have reservations that FT is suitable for this tho - from a user perspective. When zoomed out, they will see a map with 10,000 points - have to be very careful to keep that map useable. If the whole map is obscured in points, then can't see the map to zoom in! (although say an overview map may help with this) -- You received this message because you are subscribed to the Google Groups "Google Maps JavaScript API v3" group. To post to this group, send email to google-maps-js-api-v3@googlegroups.com. To unsubscribe from this group, send email to google-maps-js-api-v3+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-maps-js-api-v3?hl=en. -- You received this message because you are subscribed to the Google Groups "Google Maps JavaScript API v3" group. To post to this group, send email to google-maps-js-api-v3@googlegroups.com. To unsubscribe from this group, send email to google-maps-js-api-v3+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-maps-js-api-v3?hl=en.
