each drive has 8 io (open,close,cancel,write,read,unlink,check_callback,queue) and additional 4 for squid
running at 6 drives(COSS,AUFS per drive) 32K max/min-size.
for lowest latency we could say async-io to ((8 * 6)+4 ) = 52 ... overloading.
so i've doubled+4. now its 100 ...overloading.

can I keep increasing? up to 8 times?

--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Reply via email to