[OpenAFS-devel] RX Performance

[chas williams]

>We spent some time on this problem and found that increasing the mtu was the
>only thing that helped.  We didn't try to figure out why rx is slow, we just
>bypassed it.
>...
>NFSv4 eliminates rpc over udp completely, and requires tcp transport.  Maybe
>afs should do the same.

i dont believe tcp is the answer.  in the vafs work, rx was replaced by
aal5 for data transactions.  aal5 is more like udp than tcp.   the aal5
datagrams didnt consist of a number of standard rx datagrams.  it was
simply a biggish packet (or pdu) with an rx header.  very easy to handle.
there are some other limits in the fileserver that probably 'held back'
filserver performance.  if i remember, there is some strange 45k limit
in the fileserver.

further, use of tcp would tend to limit scalability/number of clients.