Comments on: TileCache: Map Tile Caching http://crschmidt.net/blog/archives/181/tilecache-map-tile-caching/ Ramblings of a GIS Hacker Wed, 23 May 2012 05:02:04 +0000 http://wordpress.org/?v=2.7.1 hourly 1 By: Christopher Seepe http://crschmidt.net/blog/archives/181/tilecache-map-tile-caching/comment-page-1/#comment-19628 Christopher Seepe Wed, 09 Apr 2008 14:44:24 +0000 http://crschmidt.net/blog/archives/181/tilecache-map-tile-caching/#comment-19628 You mentioned "serious services" for hosting and deploying large datasets. Could you identify a few of them for us? We have a large library of high resolution aerial orthorectified imagery (100s of TB). We have a client needed to Google Fusion pre-process perhaps 2 PB of data. Many of our customers are each in the 4 - 8 TB range. The applied research work you are doing here has substantial implications on our ability to deliver imagery for 2D and 3D online (and offline) mapping applications. Any insight and assistance would be truly and greatly appreciated. You mentioned “serious services” for hosting and deploying large datasets. Could you identify a few of them for us?

We have a large library of high resolution aerial orthorectified imagery (100s of TB). We have a client needed to Google Fusion pre-process perhaps 2 PB of data. Many of our customers are each in the 4 - 8 TB range.

The applied research work you are doing here has substantial implications on our ability to deliver imagery for 2D and 3D online (and offline) mapping applications.

Any insight and assistance would be truly and greatly appreciated.

]]> By: Chris Holmes http://crschmidt.net/blog/archives/181/tilecache-map-tile-caching/comment-page-1/#comment-10492 Chris Holmes Mon, 13 Nov 2006 16:49:21 +0000 http://crschmidt.net/blog/archives/181/tilecache-map-tile-caching/#comment-10492 Actually, <a href='http://chris.narx.net/2006/10/28/osgeo-tiling-spec/'>Chris Tweedie</a> wrote the post on S3, my post was more about using EC2 to do a batch processing job to generate all the tiles you need. Tiles for large datasets can take days if not weeks, with computing on demand from EC2 you could do it in much less time, throw a thousand servers at it. Though even better would be distributed processing of tiles, so people could volunteer processing power to create a huge tile set. Actually, Chris Tweedie wrote the post on S3, my post was more about using EC2 to do a batch processing job to generate all the tiles you need. Tiles for large datasets can take days if not weeks, with computing on demand from EC2 you could do it in much less time, throw a thousand servers at it. Though even better would be distributed processing of tiles, so people could volunteer processing power to create a huge tile set.

]]>