x
Optical/IP

Alcatel & TiMetra Seal the Deal

Today, Alcatel SA (NYSE: ALA; Paris: CGEP:PA) announced it will buy stealthy IP/MPLS edge routing startup TiMetra Networks for $150 million in stock (see Alcatel to Buy TiMetra).

Light Reading broke the story of the possible acquisition on Wednesday (see TiMetra for Sale?).

The all-stock deal is a good one for TiMetra, which hasn’t yet revealed details of its IP service edge router and has no announced customers. Venture capitalists around Silicon Valley, who had speculated about the deal before it was announced, were surprised at the final price tag.

”Wow!” writes one VC, who didn’t want to be named, in an email to Light Reading. “What a ridiculous price. I'd like that banker representing TiMetra to be on my team! He can sell ice to Eskimos!”

If TiMetra turns out to be a sled-dog, it wouldn’t be the first time Alcatel overpaid for a startup. Back in 1998, it spent $325 million to buy Gigabit Ethernet startup Packet Engines. That acquisition went sour, and the company eventually shut down the division (see Alcatel's Packet Engines Break Down). Packet Engines' former CEO Bernard Daines ended up suing Alcatel over the deal gone bad (see Alcatel Settles with Daines).

But TiMetra says its irons in the fire are heating up. Basil Alwan, TiMetra’s president and CEO, admits the startup has been stealthy about its products, but he claims there are prototypes in customer trials. Details of the edge router will be announced at Supercomm 2003 in early June, he says.

Alwan says Alcatel wasn’t the only company interested in TiMetra. He claims the startup had multiple suitors that had made serious offers, some that included substantial cash in addition to stock. Some observers speculate that Siemens AG (NYSE: SI; Frankfurt: SIE) and NEC Corp. (Nasdaq: NIPNY) were among those interested.

Alwan denies that TiMetra has been desperately shopping itself around to find a buyer. He says the company was on the verge of signing a new term sheet for its third and final round of funding, when it decided at the last minute to entertain acquisition proposals. He claims the company had raised over $50 million in cash in its first two rounds.

The first round, led by Accel Partners, Redpoint Ventures, and an unnamed company, was for $23.6 million. A year later, in 2001, the company got another $25 million plus from an unnamed strategic partner, which Alwan describes as a major supplier. The round was never announced or reported, and details are still not available.

Alcatel seems pleased with the deal. “We did our due diligence and had been researching a number of alternatives,” says Tim Krause, senior vice president of the fixed communications group at Alcatel. “We have a lot of confidence in TiMetra’s technology, and we’re confident in the customer lab evaluations.”

All 90 of TiMetra’s employees, mostly engineers, are expected to stay with the company. Basil Alwan says he will also work for Alcatel, running the TiMetra product area. But he and Tim Krause say it’s too early to talk about specifics. The deal is expected to close in the third quarter of 2003.

Krause says the TiMetra router will fit in with Alcatel's 7770 Core IP router, 7670 Routing Switch Platform IP/MPLS switch, and Omni Ethernet switches.

Specifically, carriers will be able to use the TiMetra box to provide better service-level agreements to customers, Layer 2 MPLS VPN services, and virtual private LAN services (VPLS) -- a multipoint MPLS VPN service based on Ethernet.

TiMetra has been at the forefront of the VPLS standards developments (see Virtual Private LAN Service). Vach Kompella, an engineer working at TiMetra, is one of the authors of Lasserre–V.Komplella, the most widely supported VPLS draft in the Internet Engineering Task Force (IETF) (see Kompella vs Kompella).

But VPLS has strong critics. Luca Martini, a senior architect at Level 3 Communications Inc. (Nasdaq: LVLT) and the author of the widely used Layer 2 point-to-point MPLS VPN implementation, has told Light Reading that he thinks VPLS doesn’t scale well and that he really sees no useful or practical purpose for it. (To get more, check out his interview: Luca Martini, Level 3.)

There could another strike against TiMetra -- its router won’t support broadband aggregation, Alwan acknowledges. This could be viewed as a disadvantage. Cisco Systems Inc. (Nasdaq: CSCO) and Juniper already offer B-RAS (broadband remote access server) functions on their routers. Redback Networks Inc. (Nasdaq: RBAK) is supposedly adding it to its edge router, and Laurel Networks Inc. has recently announced B-RAS support (see Laurel Joins B-RAS Pack). IP service switches like those from CoSine Communications Inc. (Nasdaq: COSN), Network Equipment Technologies Inc. (net.com) (NYSE: NWK), and Nortel Networks Corp. (NYSE/Toronto: NT) also offer it (see CoSine Unveils Next-Gen Silicon).

This acquisition seems to be part of a trend. On Tuesday, Tellabs Inc. (Nasdaq: TLAB; Frankfurt: BTLA) announced that it will acquire IP/MPLS multiservice startup Vivace Networks for $135 million (see Tellabs Snags Vivace for $135M). And last month, Ciena Corp. (Nasdaq: CIEN), which sells optical switching and transport gear, bought multiservice startup WaveSmith Networks Inc. (see Ciena Nabs WaveSmith). Two weeks ago, Lucent Technologies Inc. (NYSE: LU) announced its partnership with Juniper Networks Inc. (Nasdaq: JNPR) (see Lucent Partners With Juniper).

It’s no coincidence that all these partnerships and acquisitions are happening at once. Carriers are getting more and more serious about IP/MPLS deployments, and that’s forcing equipment vendors to quickly come up with strategies (see BellSouth Unveils MPLS Backbone).

“We’ve passed the point of IP/MPLS hype, and now it’s a reality,” says Stephen Kamman, an analyst with CIBC World Markets. “Carriers are asking for it now, but just like the transition to ATM in the early 1990s, it won’t happen overnight.”

— Marguerite Reardon, Senior Editor, Light Reading

Want a deeper understanding of MPLS? Check out the first module of Light Reading University's course on the topic. Click on this link to check it out for free!

<<   <   Page 2 / 4   >   >>
wilecoyote 12/5/2012 | 12:02:39 AM
re: Alcatel & TiMetra Seal the Deal I know what Cisco had to do to win that business which is reason number 859 million for me to hate Cisco. You know what they gave away, too, fm. They didn't win that business, they bought it. Now I know why Timetra lost the biz. Oh well, Alcatel can take Timetra a lot of places, including wireless.

Watch for SBC to announce a major reseller agreement for l-3 switching with another vendor to hold Cisco's feet to the fire and get them to drop prices. Should be fun times ahead!
firstmile 12/5/2012 | 12:02:39 AM
re: Alcatel & TiMetra Seal the Deal Wile,
No argument...but with regards to SBC.
Saw their product in the lab for EON, but Cisco won the award (by beating out Rstone). So where does that leave ALA and Timetra?
fm
firstmile 12/5/2012 | 12:02:38 AM
re: Alcatel & TiMetra Seal the Deal They definitely bought the business...reason number 981 for me :-)
fm
echo2 12/5/2012 | 12:02:34 AM
re: Alcatel & TiMetra Seal the Deal >1) Does the TiMetra buyout valuation change if
>Alcatel (ALA) stock keeps going down ?

It goes down with Alcatel, which is where it belongs.

>2) In light of this acq, do you think Procket
>is indeed worth more than $800Million,
>and the buyer will be Alcatel ?

$800Million, hahahahaha, that is too funny, I said the bubble was back, but I didnt say it was bigger than before. Alcatel wont have anything to do with them because they are 99% hype. Alcatel will only buy companies that are 100-150% hype.
GO_PHOTON 12/5/2012 | 12:02:34 AM
re: Alcatel & TiMetra Seal the Deal 1) Does the TiMetra buyout valuation change if
Alcatel (ALA) stock keeps going down ?

2) In light of this acq, do you think Procket
is indeed worth more than $800Million,
and the buyer will be Alcatel ?
Holy Grail 12/5/2012 | 12:02:27 AM
re: Alcatel & TiMetra Seal the Deal
Say what you like, but my understanding of this deal and my experience (call me a cynic if you like) makes me think that this is bum deal for Alcatel, and I still see many inherent risks..

The risks as I see them...

1) TiMetra is a risky proposition because (By their own admission in the statement made regarding the deal) TiMetra only have prototypes today, there must therfore be a real risk that the product may not in fact ever work properly or be deployable. I&#8217;ve seen this many times in the past, generally the company selling itself makes wild claims about fantastic new breakthrough hardware/price/performance/scalability, what they generally fail to point out is that massive software development and debugging and testing is still required before the box can actually be deployed in any real world IP network. I think it is the claims made about their network processor array approach and their exec background in hardware that sets my alarm bells ringing here. They may well have a neat hardware design, but there is a lot more to penetrating the carrier IP space than that.

2)Risky technology bet. TiMetra have pretty much hung their hat on VPLS, and yet it is not at all clear if VPLS will indeed become a useful technology in the carrier space, as Martini points out in his article on the LR site. I would say that he is not alone, most carriers who know what they are doing have concerns about VPLS and historically bridging type models such as VPLS and ATM LANE have suffered from scaling issues and management issues. I would say that the SP jury is out on this right now.

3) Risk missing an opportunity. Since Alcatel has such a strong position in the xDSL market, why would they not seek to acquire an edge routing company with BRAS, the fact that they have not done this surely must mean that they now risk missing an opportunity to exploit and expand their position in this important and fast growing market. Perhaps TiMetra will claim that they can buy this code in and add BRAS, and if that is the case, then again this is classic sales type BS experience shows that software development is never this easy in practice.

4) Risk that this does not deliver the IP goods &#8220;Alcatel Strengthens IP Portfolio with acquisition of TiMetra&#8221; perhaps this would better read, Alcatel starts IP Portfolio with the Acquisition of TiMetra. Anyone like to comment on TiMetra&#8217;s IP experience or their engineers track record? I seem to remember an article on LightReading back on June 14th 2002 where Alwan admits that they got their routing code from 3rd parties (see below). It&#8217;s this and their general background in hardware development and Enterprise layer 2/3 switches that makes me think that there must be a huge question mark hanging over their ability to build good IP routing code, I mean code good enough to satisfy large carriers.

And finally, if TiMetra are buying in their IP routing code then again I'm not sure how exactly this helps Alcatel strengthen their IP portfolio. By my reckoning they need all the IP software help they can get, can TiMetra deliver?

I guess only time will tell?


From LR June 14th 2002. http://www.lightreading.com/do...
While Timetra may have a lot of expertise in network processor technology, it seems to be lacking in routing software expertise. In fact, six of its eight top executives worked in the enterprise switch division of Nortel Networks Corp. (NYSE/Toronto: NT - message board) before coming to Timetra. Three of them, including Alwan, had been with Rapid City Communications, a Layer 3 Gigabit Ethernet switch startup that was bought by Bay Networks in 1997 (Nortel bought Bay Networks in 1998). None of them seems to have much experience with routing targeted at service providers.
Alwan says Timetra has built its BGP (border-gateway protocol) code from scratch and gotten other routing protocols off the shelf from third-party providers.
lastofthebohicans 12/5/2012 | 12:02:21 AM
re: Alcatel & TiMetra Seal the Deal Hi heisenberg,

Perhaps you are right about the listing being
there a while. Home grown NMS sounds like a quick-to-market idea, but all telecom/datacom uses SNMP, and maybe TL1 (Ciena). Adding snmp after the
fact (or other style proxy agent) is a contractor's dream.

I don't know where Lumos is. Take care.
-lotb
ecipo 12/5/2012 | 12:02:19 AM
re: Alcatel & TiMetra Seal the Deal Instead of entering new markets Alcatel should watch out its competitors and concentrate in what it does best. At a time when the optical networking market seems to be picking up Alcatel looks like losing market share there. Not a single contract announced there for a long time at a time when competitors are are getting everything that's up for grabs.

http://www.marconi.com/html/ne...

http://www.marconi.com/html/ne...

http://www.ciena.com/news/2003...

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

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

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

http://www.lucent.com/press/05...

http://www.lucent.com/press/05...

http://www.siemens.com/index.j...

firstmile 12/5/2012 | 12:02:19 AM
re: Alcatel & TiMetra Seal the Deal SH,
I really think that I saw the small TiMetra box and it was prior to the EON bid award to Crisco??
fm
heisenberg 12/5/2012 | 12:02:19 AM
re: Alcatel & TiMetra Seal the Deal Why should home grown NMS be a quick-to-market solution? What makes 3d party NMSs any better? In my 10-years in network management I haven't yet seen a network management package that is truly worth it. Most allow you to do very simplistic things. Most tools (all that I have encountered) are cumbersome and, in my opinion, attempt to solve all the wrong problems.. And then there are companies that claim that they can build it for you.. Have you seen their products? Rip-off is the word that comes to mind.

Another question is why should home grown NMS solution exclude SNMP? It would be just safe to assume that their box would have SNMP to the box and that their NMS would talk to the box through either SNMP or CLI or combination of both. Unless they are completely insane, they would do it this way. However, if you look at Juniper, they managed to pull it off with their XML, CLI and read-only MIBs just fine. Even if, like Juniper, they have XML interface at the box level, if it's half decently designed, it would be a matter of writing a mediation layer.. Yes, it is a contractor's dream, but it's not that difficult to do, of course, if you stay away from hiring contractors, in which case it would become money pit.
<<   <   Page 2 / 4   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE