x
Optical/IP

Nortel, Laurel Extend B-RAS Lines

Nortel Networks Ltd. (NYSE/Toronto: NT) and Laurel Networks Inc. both are releasing new broadband remote access server (B-RAS) boxes this week, advancing product lines that have gone a long time without any upgrades.

Nortel today announced a larger version of its Shasta 5000 Broadband Service Node, a release intended to push the platform into more of a mainstream B-RAS standing. Although the new box really is the latest Shasta, it's getting a new name: the Services Edge Router (SER) 5500 (see Nortel Upgrades Shasta Router).

Shasta is better known as an IP-security platform, competing with CoSine Communications Inc. (Nasdaq: COSN), rather than a serious B-RAS player, says Graham Beniston, an analyst with Heavy Reading, Light Reading's paid research service. "They've beefed up some of the processing [to enhance security features] but as a B-RAS, it needed some more session density," Beniston says.

To take care of that, the SER 5500 doubles Shasta's density, to 64,000 subscribers, while claiming the ability to run stateful firewalls for each session. The release also includes a major revision of Shasta's software.

Nortel sees services such as hosted VPNs starting to merge with B-RASs, and the SER 5500 is being positioned to tackle both areas, says Terry Boland, Nortel director of marketing. In the long run, Nortel expects B-RASs to get absorbed into edge routers. That makes the SER 5500 a stopgap, marking time until the day when the B-RAS becomes part of an edge-router platform such as the recently announced Neptune, formally named the MPE 9000 (see Neptune Arrives).

The SER 5500 is needed for now because the addition of higher-layer services into the MPE 9000 line is "further out in the rollout -- quite a bit further," Boland says. "The MPE [for now] is going to focus on Layer 2 and Layer 3 convergence and not touch on the higher-layer services." Moreover, Nortel officials have stated they intend to continue the Passport and Shasta product lines, even though both overlap with the MPE 9000.

On the customer front, Nortel has announced that Savvis Communications Corp. (Nasdaq: SVVS) is running trials with the SER 5500. General availability for the platform is slated for July.

Laurel, meanwhile, released the ST50 yesterday, its first new system since the ST200 two years ago (see Laurel Launches New BRAS and Laurel Joins B-RAS Pack). Though it's an edge router, the ST50 will likely see action as a B-RAS at first, says Steve Vogelsang, Laurel's vice president of marketing.

The ST50 is a shrunken version of the ST200, supporting 32,000 sessions versus 128,000. It was created because the ST200 is overkill for some situations. "In some portions of the network you need the features, but you don't need the number of subscribers," Vogelsang says.

The ST50 is half the physical size of the ST200, making it one-fourth of a rack, but its switching capacity is much lower -- 5 Gbit/s compared with 160 Gbit/s for the ST200. The box targets lower port speeds, with interfaces for only OC3 (155 Mbit/s) and Gigabit Ethernet. The ST200 carries those interfaces, too, but also goes up to OC192 (10 Gbit/s).

Equipment vendors commonly trot out half-sized versions of their flagship boxes (see Sources: Cisco Building 'Son of HFR'). But for Laurel, this could turn out to be a crucial step. That the company has been passed up by potential acquirers could be an indication that the ST200 wasn't what the market was looking for, Beniston says. "Laurel and some of the other B-RAS contenders have been around for a while now. The market is maturing, and you have to wonder if any of these guys are going to make it."

Laurel seemed ripe for acquisition by a larger player, but instead the deals went to others such as Vivace Networks, acquired by Tellabs Inc. (Nasdaq: TLAB; Frankfurt: BTLA), and TiMetra, acquired by Alcatel SA (NYSE: ALA; Paris: CGEP:PA). Laurel does have a large OEM partner in Ciena Corp. (Nasdaq: CIEN), which can sell a combination of Laurel and WaveSmith boxes at the edge. The lack of acquisition prospects was enough to knock Laurel off Light Reading's Top Ten Private Companies.

— Craig Matsumoto, Senior Editor, Light Reading
Page 1 / 2   >   >>
materialgirl 12/5/2012 | 1:37:41 AM
re: Nortel, Laurel Extend B-RAS Lines Can NT do Layer 3? They call this new Shasta a "Services Edge Router". Is it? They had to enter a hasty joint venture with AVCI for core routing. They bought routing technology to make Neptune. If Shasta had decent routing software, then why buy new code for Neptune? Something does not add up.
ragho 12/5/2012 | 1:37:35 AM
re: Nortel, Laurel Extend B-RAS Lines
Tom, Dick and Harry can do Layer 3 these days. The SER looks like a rebranding strategy with new CMC and SSC cards. I don't think there is anything new with fabric throughput or I/O density, but they've got a decent story there anyway.


What surprises me is the perceived redundancy in portfolio. Neptune was planned as a multi-service edge router, combining the strengths of both Shasta and Passport. Passport has needed to deliver L3 VPNs, Shasta needed to provide better BRAS, multi-service switching required MPLS so Neptune made a lot of sense. Now they have three product portfolios (Shasta/SER, Neptune and Passport) and OnlyGodKnows(TM) what the marketing strategy they plan to use for each.


Can NT do Layer 3? They call this new Shasta a "Services Edge Router". Is it? They had to enter a hasty joint venture with AVCI for core routing. They bought routing technology to make Neptune. If Shasta had decent routing software, then why buy new code for Neptune? Something does not add up.


materialgirl 12/5/2012 | 1:37:21 AM
re: Nortel, Laurel Extend B-RAS Lines Maybe Tom, Dick and Harry can do Layer 3, but can NT? Seriously. LU can't do anything these days but buy startups and try to resell their product. Does NT have the same disease when it comes to routing?

Just like CSCO's re-writing IOS implies a problem, a redundant product line from NT, including Neptune, Shasta and Passport, implies a precieved weakness in the existing products. If the problem is with Passport lacking L3 VPNs and Shasta lacking MPLS, the problem smells Layer 3-ish. If NT is too L2-ish, perhaps their products will not scale due to broadcast storms, especially with time sensitive traffic.
Mezo 12/5/2012 | 1:37:20 AM
re: Nortel, Laurel Extend B-RAS Lines Anyone can do layer 3...better watch out...Tony Li will kick your butt :]

Anyone can get a routing stack, but to make it reliable, scale, and all that...takes some real skill...most try, make claims, but the code is weak when you lab it much less production it.

Sorry, old issue...
pschurr 12/5/2012 | 1:37:06 AM
re: Nortel, Laurel Extend B-RAS Lines To answer the question directly, of course NT cannot do L3. This has been evident for years.

Their VPN solution on Passport is a joke, the upcoming MPLS capability on Passport is a joke, the L3 in the Shasta is a real mess, with all sorts of caveats on it regarding access and trunk ports, tunnels and all sorts of rubbish.

Neptune is tacked together with off-the-shelf components (confirming what NT alraedy know - they cannot do it themselves!) because they know the Shasta is dead in the L3 space. But Neptune had to come out way to early, and now they're trying to service both the existing Passport/Shasta customers as if Neptune wasn't there, and everyone else as if Neptune was the only way forward.

Correct call re Avici... NT originally had a deal with them, moved to Juniper, then tried their own (OPC), then flapped around with the new Passport, and have gone back to Avici. Long-term partner in this space? I don't think so.

Tough spot to be in.

peter
materialgirl 12/5/2012 | 1:36:59 AM
re: Nortel, Laurel Extend B-RAS Lines Thanks Peter. My issue is this: NT talks this big game about VoIP. They have this $1B "VoIP deal" with VZ. Their stock goes to $8 with a $32B market cap.

They totally lie about finances. They have this $900M Q3-03 "adjustment", then go ahead with UNAUDITED Q4 results. What a joke.

So, they lie about everything. The house-cleaning was nice. However, I still wonder what type of V0IP solution they might have. It seems to me that they have a departmental or campus solution using L2 gear.

But, how can they sell this "$1B" deal to VZ for carrier VoIP with no L3 functionality? If the annoucement is bogus, what does VZ get out of the deal? Sounding "modern" without having to really do anything because they know it won't work? Or they that dumb? Or am I missing the point?
prefer_to_lurk 12/5/2012 | 1:36:51 AM
re: Nortel, Laurel Extend B-RAS Lines materialgirl:

What does a "VoIP network" mean to you ? I get the impression that your argument is based on Nortel's lack of a core router, which (to you) implies an inability to deliver VoIP services.

To the point you've made in some other posts, the IP network is just the transport mechanism -- Nortel's VoIP solutions are based on providing the intelligence at the edge to facilitate the network application called "voice". Doesn't really matter if the core IP network is built by Avici, Cisco or Juniper so long as it provides the QoS and reliability necessary.

Or am I missing your point ?

ptl

materialgirl 12/5/2012 | 1:36:48 AM
re: Nortel, Laurel Extend B-RAS Lines Not really. It is just that if you listen to (old) NT management, they imply that they can do a carrier-grade VoIP network by themselves. It seems to me that without some L3 control mechanisms, that is not likely to scale.

When one analyst asked if the AVCI deal was related to the VZ announcement, they answered a different question, spouting platitudes like "we LIKE AVCI". The certainly did not say "YES".

This all matters to a financial person to the degree that $1B from VZ results in profits to NT. If they are just an integrator of third party stuff at a 5% mark-up, the profit profile is different than if they were getting 60% gross margins on their own gear. Of course, if a third party is about to get a big contract, that would be nice to know too.
ragho 12/5/2012 | 1:36:39 AM
re: Nortel, Laurel Extend B-RAS Lines
The real question is 'Can NT execute on a strategy?' rather than 'Can NT do Layer 3?'

There is no lack of technical expertise at Nortel when it comes to VoIP or just IP. I don't see why people continue to see Nortel's problem as one of technical inadequacy--I honestly don't believe that is true.

Their issue has always been one of execution and delivery. They've had poor assessment of certain key segments (late to realize potential), lack of committment (many scrapped and 're'-revived projects), and inability to adapt to market changes (a la capex drops). Take the circuit aggregation, B-RAS, IP Services, MSSP applications for example:
<ol>
<li>Nortel delayed, then scrapped the Versalar 15K and 25K routers which would have fit very well in circuit aggregation and BRAS markets.</li>
<li>To rectify that, Nortel bought Shasta for BRAS play. While the original vision of Shasta was awesome, it fit well in the IP Services space than BRAS. Bad market positioning, lost revenue.</li>
<li>To overcome this, some key architectural enhancements were needed for Shasta. The next-gen Shasta was conceived, aborted, resurrected and then again flushed down the toilet. This was over three years ago--they could have had a decent multi-service story with such a box.</li>
<li>They also tried to dabble in L3 VPNs and MPLS with Passport. The older Passports (6400/7400) don't have a lot of juice left, the newer ones (15K/25K) are above the price point for many applications, so that put a monkey wrench on their Passport upgrade strategy.</li>
<li>Next, when they had an opportunity to ally or even buy out Wavesmith, they stalled. Another error.</li>
<li>Then there came Neptune. This completely segmented the portfolio without giving a clear idea of what would best fit a specific application.</li>
</ol>
But this could happen anywhere in corporate America;- Nortel as a company just hasn't cleaned house. They really need to get their act together soon, but they still can.

Alliances and ad hoc partnerships deals aren't unusual. The Versalar 25K could have given them a core router, but their use of Juniper or Avici to make a sale isn't a bad thing. They just don't have any high density routing ports (OC192/10GBE).

I tend to look at it this way. If Avici had a lot to gain from the Verizon deal, they would have certainly claimed more of the brief moments of fame. They just happen to be the apple in the pig's mouth.

-ragho-
Bitman 12/5/2012 | 1:36:35 AM
re: Nortel, Laurel Extend B-RAS Lines Avici in VZN....not.
Page 1 / 2   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE