[Info-vax] Mounting NFS disk with /DATA non-default parameters

[Sum1]

On 2013-01-11 06:38:19 +0000, glen herrmannsfeldt said:

> Sum1 <not at here.com> wrote:
>>> tcpip sho ver
> 
>> HP TCP/IP Services for OpenVMS Alpha Version V5.6
>> on an AlphaServer DS10L 466 MHz running OpenVMS V8.3
> 
>> Doing a TCPIP MOUNT blah will give a default read/write size of 8192
>> according to TCPIP HELP MOUNT /DATA.  If I want to increase the size of
>> the chunks transmitted, I expect that I can issue a TCPIP MOUNT/DATA=X,
>> where X is some multiple of 512.  If I did /DATA=16384, I expect that I
>> will be transferring 16k chunks.
> 
> As well as I understand it, yes, you can do that.
> 
> Are you using UDP or TCP?
> 
>> The server and VMS client are connected on a Gigabit switch, the VMS
>> client is connecting at 100Mbit/Full (according to LANCP) with a buffer
>> size of 1500.
> 
>> Will changing NFS data size provide an improvement in transfer speeds?
>> The NFS Server is happy for me to vary the data size???.
> 
> NFS, at least the Sun version, does read ahead to keep data ready
> to send. As far as I know, there are diminishing returns with the
> increased size. There is some overhead for each block, but it isn't
> so big as to make a big difference. Also, there is increased
> latency for the whole block to arrive.
> 
> Going to jumbo packets might be more useful.
> 
> -- glen

Thanks for the useful reply.  The "dumb" switch won't support Jumbo 
packets and there is no spare slot on the DS10L.

I am trying to improve the rate at which data arrives from the server 
to the DS10L - there is a web server running and it receives requests 
for photos, most of which are 300-300K.  Increasing the transfer size 
may make a small but noticable improvement in the photo display time….?