T.R | Title | User | Personal Name | Date | Lines |
---|
888.1 | | QUIVER::WASHABAUGH | Born to be Mild | Fri Mar 05 1993 12:31 | 7 |
| I'm not an ULTRIX expert, so I cannot help you solve your problem.
However, although moving to a DEFTA is a good idea, I suspect that
whatever is keeping your performance low on the DEFZA (it should be
working much faster!) might also keep your performance low on the
DEFTA.
doug
|
888.2 | I agree, but ... | ROMTSS::CERIOLI | | Mon Mar 08 1993 03:51 | 11 |
| I agree, but hopefully moving to DEFTA will improve performances, just
because of the hardware, by a factor greater than 2, and this will suffice
to recover from this situation.
I need to know if what above is reasonable, and if DEFTA can be used on
a DECsystem 5900 running ULTRIX, regardless of whether this is supported
or not, since this is a temporary solution.
Thanks in advance.
Andrea
|
888.3 | The DEFZA can do better than ethernet | STKHLM::TORGNY | | Mon Mar 08 1993 04:14 | 36 |
| Hi,
with the performance figures you quote for the DEFZA I doubt that it's
the bottle neck. If not the DEFTA will not help. I don't know the 5900
well enough to tell if it can be the reason, compared to my figures on
the 5000/200.
Regards,
Torgny...
<<< NOTED::DISK$NOTES3:[NOTES$LIBRARY_3OF4]DNU_OSI.NOTE;1 >>>
-< DECnet/OSI for {ULTRIX,OSF/1} >-
================================================================================
Note 20.6 What about FDDI 6 of 8
STKHLM::TORGNY 16 lines 12-JUN-1991 12:08
-< TP4 faster than NSP >-
--------------------------------------------------------------------------------
Hi,
For your reference, here is a summary of the transfer rates I've noticed
between two DECstation 5000/200, connected to a DECconcentrator 500. On all
the occasions the sender run out of cpu.
TCP/IP default 18 Mbps
UDP/IP default 20 Mbps
UDP/IP without checksum 34 Mbps
DECnet default 12 Mbps
DECnet SDU size 4357, segment buffer size 65535 14 Mbps
DECnet changed from NSP to OSI transport 17 Mbps
DECnet without congestion avoidance 19 Mbps
|
888.4 | Fixed, the bottleneck was not the FDDI, but the disk. | ROMTSS::CERIOLI | | Mon Mar 08 1993 12:22 | 8 |
| Thanks for your answers.
yes, the bottleneck were the disks.
with task to task comm. I got the figures listed in .3
Ciao
Andrea.
|