NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/dcom/lans/ethernet/1736

Text File | 1992-08-29 | 1.7 KB | 37 lines

Newsgroups: comp.dcom.lans.ethernet
Path: sparky!uunet!pipex!warwick!str-ccsun!ccsun.strath.ac.uk!craa85
From: [e-mail] ( D.W.Stevenson)
Subject: Re: Ethernet Analyzer/testing devices/software
Message-ID: <[e-mail]>
Sender: [e-mail] (News account )
Nntp-Posting-Host: coll
Organization: University of Strathclyde
References: <[e-mail]> <[e-mail]> <aycn82=@quantum.on.ca>
Date: Fri, 28 Aug 1992 09:48:47 GMT
Lines: 24

In article <aycn82=@quantum.on.ca>, [e-mail] (Andrew Boyd) writes:
|> In article <[e-mail]>
|> [e-mail] ( D.W.Stevenson) writes:
|> >the PC bus is far too slow to transfer packets to main memory at 10Mbit/s
|>
|> um, this is not quite correct.  Even using the lowly 8mhz bus, a 16-bit
|> ISA card can keep up with a 10Mbit/sec throughput.


The raw bit rate is certainly fast enough to cope with 10Mbit/s, but don't
forget that the PC has to access other peripherals such as the video card
and hard disk controller over the same bus, which takes large
chunks out of the available bus bandwidth. In a previous job,
I was the engineer who designed the hardware for the Spider
Analyser (or TTC Netlens as it is now) and it just
isn't possible to do all the things that need to be done on the bus and transfer
packets at full rate of Ethernet across the bus. The achievable transfer rate
is however quite adequate for all but the most unusual of circumstances.


Dave Stevenson                 [e-mail]
Computer Centre Communications        Tel : [phone] x3461
University of Strathclyde
Glasgow, U.K.