|
Home > Archive > alt.certification.cisco > June 2003 > CCDA Basic questions?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
CCDA Basic questions?
|
|
| nheathtti 2003-06-20, 5:24 am |
| Just started to study the CCDA and have two very basic questions so please
bear with me.
Broadcasts :- Everything I read implies broadcasts can really slow a network
segment down. However once all the "hosts" are up in a particular segement
and ARP has "learnt" all the segment MAC addresses do broadcasts stop?
Obviously if a address is flushed/aged then you need another (just 1) to
restore the ARP table.
Utilisation:- What are realistic figures for a network segment utilisation.
Have seen figures saying anything above 10% is not good. Is this actually
true in pratice or is just a ploy to sell more switches/routers.
Regards
Nicki
| |
| Hansang Bae 2003-06-20, 9:24 pm |
| In article <bcupk9$j1u$1@news.boltblue.com>, nheathtti@boltblue.com
says...
> Just started to study the CCDA and have two very basic questions so please
> bear with me.
> Broadcasts :- Everything I read implies broadcasts can really slow a network
> segment down. However once all the "hosts" are up in a particular segement
> and ARP has "learnt" all the segment MAC addresses do broadcasts stop?
> Obviously if a address is flushed/aged then you need another (just 1) to
> restore the ARP table.
If you think about it, what's different about a broadcast frame than one
that's not? Broadcasts *can* slow down end users PCs because everyone
hsa to process the frame. The OS has to be notified by the NIC that a
frame came in and that the OS should look at it. When PCs were
underpowered, broadcast frames were a huge problem. These days, it's
not so bad. There are other broadcasts than just ARP so on a modern day
network, you'll always have some broadcast frames flying around.
> Utilisation:- What are realistic figures for a network segment utilisation.
> Have seen figures saying anything above 10% is not good. Is this actually
> true in pratice or is just a ploy to sell more switches/routers.
The latter.
--
hsb
"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
******************************
******************************
********
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
******************************
******************************
********
| |
| Chris Wong 2003-06-21, 1:24 am |
| The reference I used for the CCDA had 40% as the utilization number to watch
out for on Ethernet...Did it say what medium/protocol the 10% would be
applicable to?
--
Chris Wong
A+, Server+, Network+,
CCNA, CCDA, MCSA, MCSE
"nheathtti" <nheathtti@boltblue.com> wrote in message
news:bcupk9$j1u$1@news.boltblue.com...
> Just started to study the CCDA and have two very basic questions so please
> bear with me.
>
> Broadcasts :- Everything I read implies broadcasts can really slow a
network
> segment down. However once all the "hosts" are up in a particular segement
> and ARP has "learnt" all the segment MAC addresses do broadcasts stop?
> Obviously if a address is flushed/aged then you need another (just 1) to
> restore the ARP table.
>
> Utilisation:- What are realistic figures for a network segment
utilisation.
> Have seen figures saying anything above 10% is not good. Is this actually
> true in pratice or is just a ploy to sell more switches/routers.
>
> Regards
>
>
> Nicki
>
>
>
>
| |
| Who Me 2003-06-22, 4:24 pm |
| On Fri, 20 Jun 2003 10:58:59 +0100, "nheathtti"
<nheathtti@boltblue.com> wrote:
>Just started to study the CCDA and have two very basic questions so please
>bear with me.
>
>Broadcasts :- Everything I read implies broadcasts can really slow a network
>segment down. However once all the "hosts" are up in a particular segement
>and ARP has "learnt" all the segment MAC addresses do broadcasts stop?
>Obviously if a address is flushed/aged then you need another (just 1) to
>restore the ARP table.
the expiry time can be surprisingly short - I'm not sure where its set
on win or linux but you can easily check it with a small lan (2
devices + crossover cable) by pinging a device then checking the arp
cache every few seconds with "arp -a" (believe its the same in most
os's)
as mentioned elsewhere - arp broadcasts are really quite trivial even
on a huge broadcast domain.
But - anything trivial can become disasterous if taken beyond
"normal". I work in a polytechnic which, at the moment, has roughly
15,000 broadcasts a minute - 92% being IPX-RIP. The norm used to be
about 1000 frames per minute. I am convinced if the IT dept people
could find the time to find the cause and remove these unnecessary
IPX-RIP frames then our PC's would seem to speed up.
>Utilisation:- What are realistic figures for a network segment utilisation.
>Have seen figures saying anything above 10% is not good. Is this actually
>true in pratice or is just a ploy to sell more switches/routers.
If you can find real decent utilisation figures then I would LOVE to
hear them. Yes - I've heard 10% too - but think about it. If you have
an idle network and a single ARP frame is broadcast then you have 100%
utilisation by broadcast. I managed to scare our ITS when they spouted
these figures by pointing out an 8 hour period where broadcast was 89%
of the traffic between core switches. They panicd - when I told them
that this was between 11pm-7am they started to understand that you
need to consider more than just a simple percentage (ie between these
hours noone is working so no "real" traffic to compare with the
broadcasts)
All the best with your study
Dave
|
|
|
|
|