T.R | Title | User | Personal Name | Date | Lines |
---|
1175.1 | | SSDEVO::PARRIS | The SLED is dead, long live RAID | Wed Dec 08 1993 14:58 | 4 |
| This is legal and will work.
VAX systems which have an XMI bus, Q-bus, or TURBOchannel can also be connected
directly to FDDI, of course.
|
1175.2 | | KONING::KONING | Paul Koning, B-16504 | Wed Dec 08 1993 16:52 | 3 |
| ... or EISA (oh yes, that's Alpha only)...
paul
|
1175.3 | Thanks to .1 and .2 | EEMELI::SOHLMAN | Jouko Sohlman CS/PTG Helsinki Finland | Thu Dec 09 1993 01:40 | 0 |
1175.4 | More about DECbridge600 | EEMELI::SOHLMAN | Jouko Sohlman CS/PTG Helsinki Finland | Mon Dec 27 1993 07:44 | 18 |
| Hello !
Could somebody explain what does the figures about forwarding and
filtering rates mean when you talk about DECbridge620 with
three ethernets. (Forwarding rate/ethernet???)
In the materials I have the figures are:
Filtering rate = 480.000 pps
Forwarding rete = 22.000 p (Not so good if you calculate it per
ethernet == 7.400 pps/eth. !!!!)
Also posted to decnet and FDDI-notes
Jouko
|
1175.5 | 22k forwarding is enough for real-world apps | NOTAPC::LEVY | | Thu Dec 30 1993 11:45 | 12 |
| Filtering rate of ~480k packets is worst-case (all min size packets) on
all four ports.
Forwarding rate of ~22k packets is obviously less than the worst-case
rate of ~45k packets. The DECbridge was originally designed to support
one Ethernet, not three. However, 22k is completely adequate for almost
all real-world applications. Only back-to-back packets shorter than
~120 bytes (simultaneously on all three ports) would exceed the 22k
figure.
There is a white paper available which explains in much more detail
why 22k forwarding isn't a relevant performance limitation.
|