Cisco's Speculative Spec

COMMENTS

Newest First | Oldest First | Threaded View

<< < Page 2 / 2

jacksullivan66 12/5/2012 | 12:05:21 AM

re: Cisco's Speculative Spec It appears I'm not the only one who's projecting (and excited about!) a combination of the wired / wireless switch:

http://www.nwfusion.com/newsle...

To quote Joannie Wexler from her article:
"Someone will collapse a wired Ethernet switch
into the same box as a so-called wireless
switch for supporting both populations of
users from a common device. Hallelujah!"

And, from Joannie's article, it appears Airflow Networks is already working on this combination. I'd expect others are doing the same...

Reply | Post Message | MESSAGES LIST | START A BOARD

Callisch 12/5/2012 | 12:05:20 AM

re: Cisco's Speculative Spec WLAN switches already provide provide data path forwarding for both 802.11 as well as 802.3 packets....so the marriage between wired and wireless is already there.

Problem is, corporations have already made massive investments in low-cost wired switches in the wiring closet.

They aren't looking for a no-named startup to provide wired connectivity to desktop users provided by a box with no real track record.

Plus coporation today are extremely cautious about mixing their wired and wireless traffic given all the security concerns. What they want is the ability to connect their APs into existing L2/L3 switches in the wiring closet and deploy a single data center WLAN switch (with no physical or logical reconstruction of their network) to control all their APs and provide wireless services.

Reply | Post Message | MESSAGES LIST | START A BOARD

dmitton 12/5/2012 | 12:05:20 AM

re: Cisco's Speculative Spec several points:
- Joanie has been on top of this for awhile. Go back and read her columns from last couple months. Follow some of the related links backwards.

You should be watching Airespace, Aruba, Trapeze, Symbol, to name a few.

- Zorn is a prolific IETF author (dare you to scan the IETF archives and count the hits) and a friend of the primary author. Stretching his partical participation into a Cisco strategy is just too much... well, I won't go there.
Suffices to say, I read the title of this article as typical journalistic sensationalism with little true meaning.

Regardless of the bogus tie-in, Cisco is indeed in this pickle. They are invested in heavy-weight APs. I don't expect them to do anything but throw tomatoes at the vendors bucking the status-quo, until they decide to join them.

Dave.

Reply | Post Message | MESSAGES LIST | START A BOARD

jacksullivan66 12/5/2012 | 12:05:20 AM

re: Cisco's Speculative Spec "Why is this a good idea?"

Let's see:

Cost = likely less for one box (with dual functionality) when compared to two seperate boxes, likely from two seperate vendors.

Complexity = likely higher for two seperate boxes, likely from two seperate vendors.

Strategic = most enterprises are aiming longterm for one common, underlying infrastructure that can support all data types across the corporate networks - data, voice, video. Integrating the core WLAN management equipment into the core wired equipment is a logical step in that direction. Keeping them seperate requires two seperate systems, and as mentioned above, increases complexity, potential incompatibility, overhead, and cost.

Reply | Post Message | MESSAGES LIST | START A BOARD

joset01 12/5/2012 | 12:05:19 AM

re: Cisco's Speculative Spec I agree with all of that -- combining this stuff is eventually the way to go -- I think most of the startups are thinking this way as well. Here's the big problem:

- A network manager migtht buy a separate wireless switch to pilot a WLAN project in their company, that's not a huge layout and the switch can be hung off their existing network.

- But replacing your existing wired infrastructure with a multi-purpose architecture is a big outlay. I don't know if yoou've ever worked for a large utility company or another kind of corporation, but from what I rememeber the people that sign off on these kind of implemnetations are very conservative & nervous about making a mistake about a long-term update. I don't think network managers are going to buy a new architecture from a relatively untested wireless startup.

-- In fact, the longer the architectures stay separate (wired & wireless) the better it will be for the startups, because that means that network managers have more time to work with equipment and get used to it. As it stands at the moment I don't believe any network managter would decide to use a multiple purpose architecture from AirFlow (or any of the other startups for that matter) over something from Extreme or Cisco. Remember, just cos its good technology doesn't mean people will automatically buy it.

DJ Unstrung

Reply | Post Message | MESSAGES LIST | START A BOARD

jacksullivan66 12/5/2012 | 12:05:19 AM

re: Cisco's Speculative Spec Thanks for chiming in, dmitton. Agreed... I've been a reader of Joannie's column for some time now, and have been watching the emerging WLAN switch equipment space since it's early roots were established (which isn't that long ago)... ;)

Regarding "Cisco's pickle" - again agreed. They have to join the party sooner or later. Expect it to come at the hands of one of the upstart WLAN switch vendors you mentioned - via acquisition.

Regarding the title of my previous comments as "typical journalistic sensationalism" -
I simply pulled a quote from Joannie's column for emphasis - as referenced by the link. Nothing more...

Regarding Zorn's vision - I'm a fan of his IETF "thin AP" spec - so please send my regards. I'm simply sorry to see that the mother ship reigned in his very promising vision - and was trying to identify the reasons why...

Reply | Post Message | MESSAGES LIST | START A BOARD

Steve Saunders 12/5/2012 | 12:05:18 AM

re: Cisco's Speculative Spec "Problem is, corporations have already made massive investments in low-cost wired switches in the wiring closet."

This is the issue in a nutshell.

The history of LAN technology is littered with combo devices that didn't make it because they didn't integrate well with the existing infrastructure.

You could make a fridge that contained a microwave oven. Doesn't mean it's a good idea.

Reply | Post Message | MESSAGES LIST | START A BOARD

rphanley 12/5/2012 | 12:05:12 AM

re: Cisco's Speculative Spec Here you go...

http://www.nortelnetworks.com/...

Light AP, core switch, seamless roaming with security. Scaleable and manageable.

No problem mon.

1-800-4-NORTEL

Reply | Post Message | MESSAGES LIST | START A BOARD

airbb 12/5/2012 | 12:05:12 AM

re: Cisco's Speculative Spec Now, we have a few people here who understand what it means to deploy new equipment in the existing market. Add, not replace.

Also, this is why a few good startups in the wireless switch market will make it.

Reply | Post Message | MESSAGES LIST | START A BOARD

jacksullivan66 12/5/2012 | 12:05:06 AM

re: Cisco's Speculative Spec Good comments all around.

Will the WLAN switch guys sell some boxes in the short term? Absolutely! It's a great solution to many of the current problems facing today's enterprise class WLANs. But most of the big guys including Extreme and Nortel (as "vswr" was so kind to point out) are playing here already. So how big is the window of opportunity for the startups? Really?

The big question is Cisco. Because of their strong presence in the enterprise WLAN AP market, and their strong presence in the wired switch / router market - their absence from the dedicated WLAN switch movement has us all speculating on how the market will evolve - and frankly, seems to be fueling the debate...

Regarding the "dedicated equipment" vs "combination box" debate - I'm going to defend my initial comment that, ultimately, the WLAN swith functionality will be rolled up into current wired network infrastructure equipment. And possibly sooner rather than later. The key here is a belief that the enterprise is moving toward a "converged network". If in the long term the enterprise wanted two seperate networks, than I'd agree that the equipment would stay seperate. But that's not what the enterprise really wants. And it's not the vision most of us have of the future of the enterprise network.

It will ultimately be a converged infrastructure. The only real question is how long will that take?

Reply | Post Message | MESSAGES LIST | START A BOARD

<< < Page 2 / 2