off the top of my head, my rule of thumb is to check the amount of RAM required to render your largest layer which requires all the of the layer, ie at the full extents of your map. Ideally having twice the RAM available required for that will help.
SDF is also more efficient that a DB layer (ie king oracle) in this case, In my testing the same data in SDF will be processed and rendered faster which means the memory requirements are less for SDF because it turns around the request faster. ie rendering a 128MB layer from SDF might take 1s and use say 128MB of RAM, the same layer from oracle might also use 128MB of RAM but take 2s. So with 256MB of free RAM, you can serve say 2 requests with SDF per second but only 1 Oracle layer without paging. This issue http://trac.osgeo.org/mapguide/ticket/272 relating to Mapguide doing a select * from table (instead of just the required columns) also dramatically affects memory usage depending on your tables. saying all of that, when i have run up a VMWARE of XP with MGOS on my laptop it was a lot slower than running natively, but it was a laptop! z On Feb 7, 2008 10:52 AM, Martin Fafard <[EMAIL PROTECTED]> wrote: > It's a little bit slower for me. > > Martin F > > James Card a écrit : > > We have a client asking about performance of the MapGuide server on > > VMware, compared to running the server on the bare machine. I know > > some of the folks here are running the MapGuide server in VMware > > environments -- have any of you made any comparison tests? Is more RAM > > required? Does speed degrade noticeably? > > > _______________________________________________ > mapguide-users mailing list > [email protected] > http://lists.osgeo.org/mailman/listinfo/mapguide-users > -- Zac Spitzer - http://zacster.blogspot.com (My Blog) +61 405 847 168
_______________________________________________ mapguide-users mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/mapguide-users
