mherger;295334 Wrote: 
> > SC on the other hand is different and much simpler so the
> compatibility
> > tradeoff (e.g. compression vs. browsers) could be shifted to the
> user
> > and his own installation.
> 
> We did some quick tests yesterday. While gzip compression reduced the
> size  
> of our large JS/CSS collections by 75% and 85% respectively, the JS  
> download from a ReadyNAS took about 3-4x longer due to added lag for  
> compression (time for FF to download the file, measured with  
> Firebug/YSlow). Sure, once it's downloaded, it can be cached on the
> client  
> side, reducing that amount. But there's little advantage for the
> standard  
> use case (local network). I'll rather concentrate on reducing the
> number  
> of requests done to the server (currently ~100 for the start page!) by 
> 
> combining stylesheets, scripts and some of the artwork files
> (sprites).
> 
> Michael

I almost suspected those NAS-Boxes cannot be compared to dedicated
webservers. If compression would be always on i fully agree with your
conclusion. Reducing the amount of requests is certainly a good idea
with benefits to both types of server as long as you're still able to
manage those combined files.

pippin;295366 Wrote: 
> Also, some browsers cache the uncompressed data so there would not be a
> size advantage then and the penalty will hurt.

i do see the advantage of compression having better transfer times
through the wire/air not from the local disk. From my experience
reading/writing from local disks without any processing (encryption,
compression, ...) is almost always the fastest. As long as you don't
have to save disk space - makes less sense for just a few MB - it
should be okay.

kind regards,
Markus

P.S. Michael: congratulations to your 10000th post :-)


-- 
schiegl
------------------------------------------------------------------------
schiegl's Profile: http://forums.slimdevices.com/member.php?userid=6654
View this thread: http://forums.slimdevices.com/showthread.php?t=46423

_______________________________________________
beta mailing list
beta@lists.slimdevices.com
http://lists.slimdevices.com/lists/listinfo/beta

Reply via email to