Re: [hybi] Framing Take VI (a compromise proposal)

[Willy Tarreau <[e-mail]>]

[ a bit off-topic ]

On Sat, Aug 14, 2010 at 02:58:48PM +0900, Takeshi Yoshino wrote:
> > But let me provide some metric about this. As you said, people who mind
> > latency much would turn off NAGLE, so short packets containing, e.g. key
> > stroke of 1 octet would ultimately forms minimum_gap(8) + ethernet(18) +
> > ipv4(20) + tcp(20) + type(1) + len(1) + payload(1) = 69 octet on wire.
> > difference of 7 octets is just 10% increase on wire and
> >
> 
> Pardon me. minimum_gap 8 -> 12 (most modern NICs work with 8, i think, but
> takes number on the 802.3). And, I missed preamble, so s/69 octet/81 octet/.
> 7/81=8.64% increase on occupation on ethernet baseband signal.

The network overhead should be considered at IP level because that's what
most network providers will count. So your 7 more bytes are 50 vs 43, which
means 16%. That said, it's hard to believe that some networks will be loaded
with 1-byte frames to the point of being significant enough. The packet
processing is much more of a concern there.

Regards,
Willy