| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1163.1 | No Performance Degradation with filters | NACAD::KRISHNA | | Tue Jun 28 1994 10:57 | 12 |
| Hi Faraday,
There will be no impact to the performance of the DECbridge
900MX when filtering (address and/or protocol) is turned on. The design
is such that for forwarding a packet with or without filters, the
lookups needed are the same and hence the cost incurred in making the
forwarding decision is the same.
This is certainly an important fact that should be used to our
advantage in selling the DECbridge 900.
Hope this helps,
Krishna
|
| 1163.2 | Mention too that Scott Bradner's tests support this too... | NACAD2::BATTERSBY | | Tue Jun 28 1994 12:18 | 6 |
| We should also use to our advantage the results of the Scott
Bradner's Lab testing which also, of course supports the fact
that there is no impact on the DB900MX Bridge performance when
filtering for protocol or address is turned on.
Bob
|
| 1163.3 | Where is Scott Bradner report? | HGRNCC::FARADAYCHONG | Faraday Chong@hgo 852-8053590 | Tue Jun 28 1994 12:59 | 7 |
| Where can I find the Scott Bradner report on DECbridge 900MX? I
couldn't find it when I checked the Internet repository a few weeks
back.
Is it on EASYnet?
Faraday
|
| 1163.4 | | NWD002::KOPEC_ST | Squash:Racketball::Chess:Checkers | Thu Aug 25 1994 11:31 | 5 |
| same question. one of our customers (who presently is considering
Digital hubs and Chipcom hubs) had heard of this report and requested a
copy. It could be the difference between a win and a loss.
Thank You, Stan
|
| 1163.5 | Scott Bradners Internet access for test reports... | NACAD2::BATTERSBY | | Thu Aug 25 1994 13:25 | 15 |
| The report on DECbridge 900MX is freely-available on Scott Bradner's
Internet repository. ftp into hsdndev.harvard.edu, directory of
interest
is /pub/ndtl/results/vendors/Digital/DECbridge_900MX.
The key performance figures he came up with are: forwarding max rate
62,488 pps; max latency from Ethernet-Ethernet 0; max latency from
FDDI to Ethernet 70 microseconds. Additional detail is in note
716 of HUB_MGNT conference.
Also read Anil's summation of DD900MX features in note 716.
[thanks Anil for forwarding me the Internet pointer to Scott Bradners
repository to post in here - Bob]
Bob
|
| 1163.6 | | NACAD::ANIL | | Wed Sep 07 1994 19:58 | 5 |
| Bradner does his test without using any management filters. Which
is not to say that the results would have been any different if
he did.
Anil
|