|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jim Henderson wrote:
> I dunno, but I know at work they just implemented a bunch of L4 switches,
> and the technology has been around for some time...
Unless you're routing between the OSI stack and TCP or something, I just
can't imagine it isn't some marketing-speak for "application-layer
gateway" or something.
--
Darren New / San Diego, CA, USA (PST)
It's not feature creep if you put it
at the end and adjust the release date.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Sat, 12 Jan 2008 17:31:45 -0800, Darren New wrote:
> Jim Henderson wrote:
>> I dunno, but I know at work they just implemented a bunch of L4
>> switches, and the technology has been around for some time...
>
> Unless you're routing between the OSI stack and TCP or something, I just
> can't imagine it isn't some marketing-speak for "application-layer
> gateway" or something.
http://en.wikipedia.org/wiki/Layer_4_switch
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jim Henderson wrote:
> http://en.wikipedia.org/wiki/Layer_4_switch
Oh. Cool. I guessed right. :-)
--
Darren New / San Diego, CA, USA (PST)
It's not feature creep if you put it
at the end and adjust the release date.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Sun, 13 Jan 2008 09:23:19 -0800, Darren New wrote:
> Jim Henderson wrote:
>> http://en.wikipedia.org/wiki/Layer_4_switch
>
> Oh. Cool. I guessed right. :-)
:-)
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>> They want everybody to use the same brand and model of... well,
>> everything actually.
>
> You know, that's not such a bad policy (when used reasonably).
No, it's not really.
Or rather, it's a good policy if the thing you choose to standardise on
is a good choice. In this case, apart from the astronomical cost,
choosing Cisco is a pretty sensible choice (if a little overkill).
> ) What can this equipment do that you currently cannot do with your
> network?
I'm guessing it's possible to configure these routers so you can monitor
traffic from any node. [Not that I will be given the tools necessary to
do this.]
It should also be easier to set up strange network configurations
temporarily if the need arrises. [Again, I won't be allowed to do this.]
> ) What might change over the next five years to make this equipment a
> good choice?
Hopefully I won't be working here long enough to care...
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
And lo on Sat, 12 Jan 2008 00:35:13 -0000, Jim Henderson
<nos### [at] nospamcom> did spake, saying:
> On Fri, 11 Jan 2008 15:07:55 +0000, Invisible wrote:
>
>> The point is it provides wildly more
>> functionallity than we actually need.
>
> Today, but when you spend money on networking infrastructure, you also
> plan for what you might need in the future. Some places just buy the
> biggest thing they can so it's not going to be obsoleted before the
> depreciation timeframe expires.
Or from the more practical view they needed 35, but get a discount if they
buy 50; so they justified the extra 15 by sending them to the UK :-)
--
Phil Cook
--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Mon, 14 Jan 2008 10:01:42 +0000, Phil Cook wrote:
> And lo on Sat, 12 Jan 2008 00:35:13 -0000, Jim Henderson
> <nos### [at] nospamcom> did spake, saying:
>
>> On Fri, 11 Jan 2008 15:07:55 +0000, Invisible wrote:
>>
>>> The point is it provides wildly more
>>> functionallity than we actually need.
>>
>> Today, but when you spend money on networking infrastructure, you also
>> plan for what you might need in the future. Some places just buy the
>> biggest thing they can so it's not going to be obsoleted before the
>> depreciation timeframe expires.
>
> Or from the more practical view they needed 35, but get a discount if
> they buy 50; so they justified the extra 15 by sending them to the UK
> :-)
That's also true. :-)
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Invisible wrote:
> OK, so the company I work for is moving to a new building.
Finally. That's one thing you've been waiting for ;).
> As part of the move, I thought it would be nice to buy some new gigabit
> brand you buy. But when HQ got wind of this, they said "oh no, you must
> buy these Cisco switches, that way they'll match what everybody else has".
Yes. Keep the line clear, and the network will be easier to maintain.
Yes. Cisco has a price. Cisco also has IOS, functionality, support,
quality and reputation. Practically Cisco is always a safe choice, when
it comes to network devices.
> Well anyway, HQ bought the switches themselves. I got an email the other
> day to say they've finished configuring them now. (It's a switch? What's
> to configure? It's a passive component...) They're about to ship them over.
No, it's not a passive component. Even the stupidest switches have some
intelligence and do active job. Manageable switches do much more and
routing switches (like the ones you'll get) are 'bout blessing for
network infrastructures.
> Yesterday I got a document describing the configuration of the switches.
> And now it all becomes horribly clear.
>
> These "switches" are actually *routers*. That's why they're so damn
> expensive - each one is a 24-port *router*!!
They are freaking reliable routing switches with world-reputation
support. Yes, that costs money.
> Um... we don't *need* routers. We have 1 subnet. We just need some
> ordinary switches. Oh well...
>
> But wait! Looking at the configuration details, it seems HQ want to
> split my network into several seperate subnets, and have configured the
> routers to route between them.
Because you can :).
> Er... why?? This isn't necessary. All this does is massively increase
> the complexity of my network. For no gain. Why are you going this?? (And
> why is today the first I've heard of this?)
It makes the network more controllable and logical (or to be precise, it
makes getting the network more controllable and logical possible). And
it increases security, if made correctly.
> *sigh* Clearly I'm going to have to make some phone calls... :-(
>
> [Seriously. Do these people just enjoy making things complicated for the
> fun of it? Are they trying to prove how cleaver they are or something?
> Perhaps this is job security? I don't know, but where I'm from, it's
> usual to go with the *simplest* solution that does what you want, not
> the most complicated one...]
They are making theier job (and yours too) more stable. It's a bit more
work to configure the system and some work to maintain it, but it
reduces big problems.
And yes, I'd love to get Catalyst to be the base of my homenet. But
because of the price I'll probably have to just get 1800-series Procurve.
--
Eero "Aero" Ahonen
http://www.zbxt.net
aer### [at] removethiszbxtnetinvalid
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Mon, 14 Jan 2008 23:12:39 +0200, Eero Ahonen wrote:
> It makes the network more controllable and logical (or to be precise, it
> makes getting the network more controllable and logical possible). And
> it increases security, if made correctly.
I don't know that I'd go with the security angle, unless 802.11x is being
implemented to authenticate devices to the network.
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>
> Yes. Cisco has a price. Cisco also has IOS, functionality, support,
> quality and reputation. Practically Cisco is always a safe choice, when
> it comes to network devices.
Oh, sure, nobody can deny Cisco is the best brand on the market. ;-)
Maybe that will become useful in the future; at the moment it's pure
overkill.
> They are freaking reliable routing switches with world-reputation
> support. Yes, that costs money.
That's true - but it's also true that the switches we've already got
have run for 10 years without one single reliability issue.
>> Er... why?? This isn't necessary. All this does is massively increase
>> the complexity of my network. For no gain. Why are you going this?? (And
>> why is today the first I've heard of this?)
>
> It makes the network more controllable and logical (or to be precise, it
> makes getting the network more controllable and logical possible). And
> it increases security, if made correctly.
I don't see it.
I mean, if we had one group of nodes that talk to each other and don't
talk to anything else much, putting them onto a seperate subnet would
make a lot of sense. But that isn't the case. We have 50 PCs and 4
servers. All 50 PCs talk to the same 4 servers and the Internet. I fail
to see how subnetting does *anything* in this situation other than
adding unecessary complexity.
> They are making theier job (and yours too) more stable. It's a bit more
> work to configure the system and some work to maintain it, but it
> reduces big problems.
Such as?
> And yes, I'd love to get Catalyst to be the base of my homenet. But
> because of the price I'll probably have to just get 1800-series Procurve.
Your home network must be *much* bigger than mine. ;-)
[Mine has 3 nodes.]
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|