[Gluster-devel] how best to set up for performance?

[Niall Dalton]

Hi,

On Mar 16, 2008, at 12:46 PM, Anand Avati wrote:
>  You might want to try specifying
> 'option window-size 2097152' in all your protocol (client and  
> server) volumes.
> Also, you might also want read-ahead on the server (over iothreads)  
> with the same page-size as the client, but higher page-count.


It seems that I'm not at all diskbound. When I repeatedly dd a 10GB  
from from the 16GB memory of one of my storage servers I consistently  
max out at around 390MB/s with _zero_ disk io involved. I imagine I'm  
doing something wrong in terms of juggling cache sizes, io threads,  
readahead sizes etc that is causing this? Or is there some other OS  
interaction interfering?

thanks
niall