So Extremely Sorry, Startups

The man running Extreme Networks Inc.'s (Nasdaq: EXTR) wireless LAN switch project says he feels sorry for the startups in the market now that the major vendors are announcing products.

As expected, Extreme announced its Summit 300 48-port Layer 2/3 switch today, configured to support both wireline and wireless applications, and a companion Altitude 300 "lightweight" 802.11 access point today, along with associated network configuration and management software (see Extreme Hatches Switch Surprise). The products are expected to ship next quarter.

As previously reported, the box follows the standard "smart media hub" approach for implementing a managed wireless LAN network in a corporate environment (For more on the debate over what exactly comprises a wireless LAN switch, see Vivato's Switch Bitch).

The entry of Extreme and other major vendors into the market (see Nortel Preps 'Security Switch') means that many of the products that the many startups in the market are expecting to ship in the next few months will be dead on arrival, according to Vipin Jain, VP and general manager for Extreme’s LAN Access Business. "I feel sorry for the startups," he says. "Customers don't want to go to a new player for this type of box." [Ed. note: And no one feels worse about it than he does...]

Chris Kozup, analyst at the Meta Group Inc. agrees: "I've always said that people will buy from established vendors rather than startups."

However, some startups may have a chance to win business due to their technological edge over the incumbents on the wireless side of things. Example: The Extreme switch does not support functions such as roaming among access points out of the box, meaning customers will have to buy a separate software upgrade to enable those capabilities.

"It looks as if there is a slight lag there in terms of the features being offered by some of the more aggressive vendors such as Aruba Networks Inc. and Trapeze Networks Inc.," says Kozup. "But it is not enough to keep Extreme out of the market." Extreme is charging $125 a port for the switch. Kozup reckons that with "Extreme's standard 25 percent discount" enterprises could be paying a "competitive" $100 per port for the switch. In five years time, the cost of offering wireless capabilities will be on a par with wired Kozup says.

Extreme is also planning to offer standalone access points next year. And Jain says the company will also look at supporting new radio technologies such as ultrawideband (UWB) when they come of age.

However, just as Jain expects the startups to be looking over their shoulder at the Extreme launch, he in turn is awaiting Cisco Systems Inc.'s (Nasdaq: CSCO) entry into the market. "This is just a fact of life," he says. "This is the only company we've ever competed with." (See Cisco’s LAN Switch: Build or Buy?.) — Dan Jones, Senior Editor, Unstrung

Page 1 / 2   >   >>
deepciscothroat 12/5/2012 | 12:13:50 AM
re: So Extremely Sorry, Startups What a effing bonehead. He must have forgotten that we once made the same claims against Extreme. Extreme's revenue is going the wrong way, so maybe they are turning into a start-up again

From what we hear, Extreme's entire effort is a kluge and they are going to look EXTREMELY stupid compared to other players. It was a shit effort inside Extreme that no one wanted to work on, but customers started complainign about. Jain is a storage monkey who knows nothing about wireless (and a lot about shooting-off his mouth)

From what we hear on Tasman Drive, Airespace and Aruba have the most customer traction, and Trapeze is starting to make a lot of noise. The Extreme wireless switch is a figment of their imagination -- where they are still great (clearly not in the marketplace)

We love when idiots like Jain make us look like innovators.


Our prediction: us vs. the start-ups. Extreme never plays in wireless. What LOSERS!!!!!

noitall 12/5/2012 | 12:13:49 AM
re: So Extremely Sorry, Startups Tell us how you really feel, CW.

The Iraqi Information Minister has surfaced!

I agree. This guy is a world class boner. He was hawking gigE and spending hundreds of millions of VC dollars after being part of the team that ruined 3Com. Now he's a deer in headlights at Extreme calling Cisco the only competition when Foundry Networks is killing them, and Cisco's only letting them live to keep the Feds out of their shorts.

"We'll burn them in their tanks. No one else is in the LAN switching market."

Nice track record for a bold statement proclaiming victory while the Marines are swarming all over them.

He has the shape of an L on his forehead.

Semper Fi, Vippin!

God, I love a good laugh these days. Too rare. Thanks Vippin!
deepciscothroat 12/5/2012 | 12:13:46 AM
re: So Extremely Sorry, Startups Hey Bleu Tooth
Heard Gordon Stitt's broadcast
It gets better...
Quote of the day. "This technology was built to collect dust"

No kidding. He said that

pearljammed 12/5/2012 | 12:13:45 AM
re: So Extremely Sorry, Startups This banter is not even modestly useful
What is the advantage of using a wireless or a wired and wireless approach for this market

Can a startup hit escape velocity?
deepciscothroat 12/5/2012 | 12:13:43 AM
re: So Extremely Sorry, Startups Knew Alan when he was @ Cisco.
007 12/5/2012 | 12:13:43 AM
re: So Extremely Sorry, Startups Sounds like you are Alan S Cohen (ASC)
the VP of marketing at Airespace.
lrmobile_emma 12/5/2012 | 12:13:42 AM
re: So Extremely Sorry, Startups Extreme seems to be missing the boat on security! They just have a switch with power over ethernet and some software and an overweight access point that does the encryption. It's no better than Cisco's offering of a 3550 Catalyst with Power over Ethernet and their sweet 1200 dual band APs. When you combine Cisco's many, many years of wireless experience and incredible AP discounts, there's no reason not to go with Cisco over Extreme. Or Trapeze for that matter. Trapeze is selling the same junk as Extreme.

However, if you take a look at some of the other startups, they are onto something, because they do encryption in the switch. That makes sense because they can support IPSec VPN termination on the switch, and the access points are cheap. I sure wouldn't want to have to forklift $600-800 access points going from 802.11b to 802.11a and then again to UWB or whatever the next PHY 100Mbps+ standard is going to be. That's like doing the yearly Cisco AP upgrade. I think it would be better to only spend $200 per AP to do that upgrade with a switch that supports centralized encryption - assuming that they can hit 200Mbps+ throughput.

wirelesstaker 12/5/2012 | 12:13:41 AM
re: So Extremely Sorry, Startups Emma
this is an interesting analysis. Why would anyone want to terminate a VPN on one of these wireless/wireline switching when there is likely to be an VPN server in the enterprise?

This identity-networking seems pretty cool and could open up some interesting possibilities

maxhead 12/5/2012 | 12:13:35 AM
re: So Extremely Sorry, Startups Let's take it point by point:

"and an overweight access point that does the encryption."

-what constitutes an "overweight" access point? The only two things I can see are hardware encryption (good) and the radio (mandatory). "Fat" AP solutions (i.e., Cisco) include local console port, and additional HW+SW.

"When you combine Cisco's many, many years of wireless experience"

-Cisco purchased Aironet in 1999--their mobile security solution is "Mobile IP". Do a goodle search on "Mobile IP problems", and tell me if this is consistant with a vendor that has "years of wireless experience". Cisco's forte is IP and the focus has been on stationary users behind physical ports. Mobility isn't built into IOS, and running it on an AP fails to convince me (and the 10s of people I've spoken to about the Cisco solution) it's the correct approach.

"Trapeze is selling the same junk as Extreme."

-Similar architecture regarding AAA and the AP switch functionality, but where are Extreme's site planning and deployment tools? Also, Extreme will likely only sell a wireless solution to their existing customer base. At $90m per quarter, itt's not a big market share. Startups like Trapeze have a market potential of every wired install base. Small things also bother me, like painting the APs purple--clearly the marketing guys won vs. the Security guys...this AP will be a prime target for malicious users. They won't get any information from the AP(good answer), but it makes me question how serious their focus on security is.

"However, if you take a look at some of the other startups, they are onto something, because they do encryption in the switch"

-VPNs are designed for remote access across a foreign network. WLAN in the enterprise is LOCAL access, and doesn't need the over-engineering that VPNs require. Also--where is multicast, QoS (for Wireless VoIP) support in VPNs? What about maintaining the same VLAN & ACL definitions the user already has defined in the wired netowrk? Simply stated, VPNs are NOT the solution for enterprise WLANs.

"I sure wouldn't want to have to forklift $600-800 access points going from 802.11b to 802.11a and then again to UWB or whatever the next PHY 100Mbps+ standard is going to be."

-How much cheaper are these "cheap" APs? The AP needs power and a radio at a minimum. Good AP solutions should include hardware accelerated encryption (Trapeze & Extreme for example) and redundancy. The same upgrade argument was made for 10/100/1000 NICs in PCs, but at the end of the day, the IT dept. will have to roll out a WLAN service, and better to have a migration strategy (dual-band radios)than to sit on the fence and do nothing while your users keep buying Linksys boxes and connecting them to their cubicle's switch port.

-Max Headroom

P.S. More practical engineering responses to vendor comparisons and less off-the-cuff bluster is welcome...
jacksullivan66 12/5/2012 | 12:13:35 AM
re: So Extremely Sorry, Startups Has anybody seen the news out today on Engim?:


After two years of stealth development, and backing by Bessemer and Matrix Partners, these guys have announced the product availability of a "multi-channel WLAN switching chipset" that "allows simultaneous communications on multiple 802.11 channels, in multiple bands,
from a single access point"! That means all channels are open and can be active, on 802.11a/b/g, all simultaneously - providing aggregate capacity of 594Mb/s per AP! And they work with any type of client/end user device


What's interesting here is now that the bulk of the switching intelligence is reduced to silicon, why do we need "dedicated" equipment like Access Points? If I were Dell or HP, I'd be putting an Engim chip inside every one of my enterprise class PCs.

With so much of the network intelligence now available in silicon, and an already tangible trend toward "dumb" APs, I'm not sure I'd want to be a WLAN equipment vendor right now. So Sorry, Extreme... Tough going Cisco...
Page 1 / 2   >   >>
Sign In