On Thu, 27 Jul 2000, Mooney Drew-DMOONEY1 wrote:
> Marc,
>
> I'm not sure I understand this.
>
> Let's say I have a hubbed network with 20 hosts. Lots of collisions.
>
> I bring a 4 port switch into it, and sub-divide it into 4 hubbed segments of
> 5 hosts each. No collisions? Or just extremely low likelihood of collisions
> on a given segment?
You will get collisions within each of the 4 collision domains you have
created.
>
> Extend the scenario - those 4 hubbed segments grow into 4 hubbed segments of
> approximately 20 hosts each. Lots of collisions on each segment but no
> collisions across segments? Or no collisions on any segment because it's
> switched?
You will get more collisions. You will not get collisions in the swithing
fabric itself, collisions will be isolated to the individule collision
domains you have created.
>
> I *thought* [don't know] the idea behind subdividing networks with switches
> was to reduce the likelihood of collisions by introducing more and more
> smaller network segments, and that as the number of hosts grows on a
> particular segment, one needs to continuously subdivide to minimize the
> likelihood of collisions.
Today, switches are cheap (compared to long ago). It is common to give
each machine its own dedicated switch port, in which case you will cut
down on collisions drastically. If you run FDX on those ports, you will
have no collisons.
>
> You and others in this thread seem to be saying that incorporating switches
> into a network is to completely eliminate collisions. True? False?
FDX/switched = no collisons
HDX/switched = some collisons (on the wire between the nic and the switch
port)
if you just hang a hub off a switch port, then yes you can get collisons
on the machine attached to that hub, and you cannot run FDX to the hub
either.
Brian
>
> Please clear this up for me....thanks.
>
>
> Drew M. Mooney
> Invisix -- Motorola and Cisco Together
> 1334-394 The Alameda // San Jose, CA 95126
> 408-525-0873 [office] 408-287-3188 [home]
> 817-937-7880 [mobile] 888-809-9678 [SkyTel Pager]
> +44-(0)7715-055-944 UK Mobile
>
>
> -----Original Message-----
> From: Marc Quibell [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, July 27, 2000 4:49 PM
> To: Drew Mooney-DMOONEY1
> Subject: RE: CCDA question-512 bit times
>
>
> I agree. Once you get out of the switched environmet, collisions will
> occur...
>
> Marc
>
>
> >From: Mooney Drew-DMOONEY1 <[EMAIL PROTECTED]>
> >To: "'Marc Quibell'" <[EMAIL PROTECTED]>
> >Subject: RE: CCDA question-512 bit times
> >Date: Thu, 27 Jul 2000 11:25:51 -0500
> >MIME-Version: 1.0
> >Received: from [144.189.100.103] by hotmail.com (3.2) with ESMTP id
> >MHotMailBB49AD91008AD820F3C690BD646794EF0; Thu Jul 27 09:25:54 2000
> >Received: [from mothost.mot.com (mothost.mot.com [129.188.137.101]) by
> >motgate3.mot.com (motgate3 2.1) with ESMTP id JAA11522 for
> ><[EMAIL PROTECTED]>; Thu, 27 Jul 2000 09:24:27 -0700 (MST)]
> >Received: [from tx14exm02.ftw.mot.com ([178.1.100.242]) by mothost.mot.com
> >(MOT-mothost 2.0) with ESMTP id JAA16879 for <[EMAIL PROTECTED]>; Thu,
> >27 Jul 2000 09:25:52 -0700 (MST)]
> >Received: by tx14exm02.ftw.mot.com with Internet Mail Service
> >(5.5.2650.21)id <PMH3ZDGN>; Thu, 27 Jul 2000 11:25:51 -0500
> >From [EMAIL PROTECTED] Thu Jul 27 09:29:40 2000
> >Message-ID:
> ><[EMAIL PROTECTED]>
> >X-Mailer: Internet Mail Service (5.5.2650.21)
> >
> >Ahhh - but suppose the switched segment is not to a single PC - but to a
> >hubbed group of PC's. The switch in that case, minimizes the likelihood of
> >collision, but won't entirely eliminate it.
> >
> >Two stations in the same segment are still able to attempt transmission
> >simultaneously.
> >
> >Drew M. Mooney
> >Invisix -- Motorola and Cisco Together
> >1334-394 The Alameda // San Jose, CA 95126
> >408-525-0873 [office] 408-287-3188 [home]
> >817-937-7880 [mobile] 888-809-9678 [SkyTel Pager]
> >+44-(0)7715-055-944 UK Mobile
> >
> >
> >-----Original Message-----
> >From: Marc Quibell [mailto:[EMAIL PROTECTED]]
> >Sent: Thursday, July 27, 2000 8:51 AM
> >To: [EMAIL PROTECTED]
> >Subject: Re: CCDA question-512 bit times
> >
> >
> >I think she answers in her book, in a roundabout way, that in FULL-DUPLEX
> >mode, collisions are non-existant, since two stations can transmit at the
> >same time on the wire (a switch and the PC or device on it's port,
> >transmitting and receiving at the same time)
> >
> >Marc
> >
> >
> >"Steve Brokaw" <[EMAIL PROTECTED]> wrote in message
> >[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> > > Well, I have a different question that kinda goes along here. If you
> >are
> >in a switched environment, i.e. dedicated bandwidth per port, how can you
> >have a collision at all? To me it seems (and Radia Perlmann touches on
> >this
> >in her book but doesn't give any explanation) that if there is no chance
> >for
> >a collision (switched environment) then why a distance limitation? I'm
> >sure
> >there are some other physics factors that would limit distance but would
> >they be the same as the distance required to detect a collision?
> > >
> > >
> > > Steve
> > >
> > >
> > > ------Original Message------
> > > From: "Randy Witt" <[EMAIL PROTECTED]>
> > > To: [EMAIL PROTECTED]
> > > Sent: July 27, 2000 1:53:55 PM GMT
> > > Subject: CCDA question-512 bit times
> > >
> > >
> > > I have a question regarding the round-trip propagation delay on an
> >Ethernet network.
> > >
> > > Page 123 of the Cisco Press "Designing Cisco Networks" book states:
> > >
> > > "The most significant design rule for Ethernet is that the round-trip
> >propagation delay in one collision domain must not exceed 512 bit times,
> >which is a requirement for collision detection to work correctly."
> > >
> > > With 100Mbps Ethernet, the maximum round-trip delay would be 5.12
> >seconds,
> >resulting in a distance limitation of 205 meters.
> > >
> > > I currently oversee a large flat network covering several miles in
> >diameter. All of the links between buildings are single-mode fiber links.
> >No routing is involved, everything is switched - one large broadcast
> >domain.
> > >
> > > How does the 512 bit time rule apply to fiber optic cabling? I see on
> >page 127 of the same book that the Round trip delay in bit times per meter
> >for Cat5 cable is 1.112, whereas Fiber-optic cable it's 1.0.
> > >
> > > I guess I'm having difficulty understanding how fiber can overcome the
> >512
> >bit-time rule and can have a much longer distance.
> > >
> > > I do realize that this is not exactly a Cisco question, though covered
> >on
> >the DCN/CCDA material. If someone could kindly refer me to any material
> >that covers this topic, I'd appreciate it.
> > >
> > > ___________________________________
> > > UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
> > > FAQ, list archives, and subscription info: http://www.groupstudy.com
> > > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> > > Steve Brokaw, MCSE CCNA
> > > Sprint Enterprise Network Services
> > > [EMAIL PROTECTED]
> > > [EMAIL PROTECTED] (pager)
> > >
> > >
> > >
> >***************************************************************************
> *
> >***
> > >
> > > Never mistake motion for action.
> > >
> > > -- Ernest Hemingway
> > >
> >***************************************************************************
> *
> >***
> > >
> > >
> > > ___________________________________
> > > UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
> > > FAQ, list archives, and subscription info: http://www.groupstudy.com
> > > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> > > ---
> >
> >
> >___________________________________
> >UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
> >FAQ, list archives, and subscription info: http://www.groupstudy.com
> >Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
> ________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
>
> ___________________________________
> UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
> FAQ, list archives, and subscription info: http://www.groupstudy.com
> Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
-----------------------------------------------
Brian Feeny, CCNA, CCDA [EMAIL PROTECTED]
Network Administrator
ShreveNet Inc. (ASN 11881)
___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
