x
Optical/IP

Cisco Eyeing Sonus

Several sources say that Cisco Systems Inc. (Nasdaq: CSCO) and Sonus Networks Inc. (Nasdaq: SONS) have been talking – and those discussions could lead to acquisition by Cisco.

So far the relationship has been described as exploratory, say sources close to investment banking interests. Light Reading could not confirm whether a deal had been made or what the price would be.

Sonus builds voice over IP (VOIP) gateways for carriers that allow them to carry voice traffic over packet-based network.

One venture capitalist familiar with the market, speaking under condition of anonymity, said a deal would likely get done at a slight premium and said he could see Sonus being sold for $5 or $6 per share. Sonus was trading at $3.20 during the late afternoon today.

Another source pointed to Paul Ferri, partner at Matrix Partners and Sonus board member, who is known for his impatience when companies hit rough patches, as Sonus has. This source tells Light Reading that Ferri may have an itchy trigger finger: "When Ferri decides to do something, he makes it happen." (See Top Ten Movers and Shakers, No. 5.)

Would a combination of Cisco and Sonus make sense? On many levels, it seems like a good fit. Here’s why:

  • Sonus is widely regarded as a leader in the voice softswitch market, and Cisco has struggled to build a leadership position in that market.
  • The capital-spending crunch in the telecom market has put the emerging VOIP market on ice, chilling Sonus’s prospects for near-term growth.
  • Sonus’s numbers have recently hit a rough patch. For the fourth quarter of 2001, the company reported revenues of $38.9 million and an adjusted net loss of $7.7 million. The company also warned that revenue was expected to decline again in the first quarter of 2002, but it has not provided specific guidance.
  • With losses, Sonus’s financial position has deteriorated, and they may be in need of a deep-pocketed partner to get them through the lean years. For example, as of the end of 2001, Sonus had $125 million in cash and short-term investments. One quarter earlier, the company held $137.4 million in cash and short-term investments.
Some analysts describe the combination as an evergreen rumor on Wall Street, but they say that's because the deal makes sense.

"It would be a great fit, because Sonus has strong technology and it's successfully deployed in large carrier networks," says Richard Church, a senior analyst with Wachovia Securities. "Cisco has talked about this as a tornado market, but so far they're only in the enterprise [market]."

Church said that with $125 million in cash, Sonus could stick it out as an independent company and have enough funds to last well over a year if it continues losing money at the current pace. However, he noted that it is more difficult to sell products to large carriers, especially RBOCs and ILECs, as a small independent company.

However, news of the talks was greeted with skepticism by some. "That rumor comes around every week," says Sam Wilson, an analyst with Merrill Lynch & Co. Inc. "But I doubt Cisco is very interested. Sonus is going after the Internet offload market, and that market has pretty much peaked already. It's also unclear if they'll offer class-4 or class-5 replacement." Wilson said Cisco is more likely to be interested in a smaller, private company.

Another factor: Cisco has said it still plans to do acquisitions this year, just on a smaller scale. But it has also said it’s looking for a company with shipping product. Sonus fits that description.

Sonus officials could not be reached at press time. A Cisco spokeswoman said the company had no comment.

— R. Scott Raynovich, US Editor, and Marguerite Reardon, Senior Editor Light Reading
http://www.lightreading.com
Page 1 / 9   >   >>
Wowwo2 12/4/2012 | 10:46:08 PM
re: Cisco Eyeing Sonus You are the documented liar !!

http://www.lightreading.com/bo...

"ADIR recently completed their VOXIS management software and gatekeeper and CSCO although not announcing it publicly has incorporated it into their most important gateways.
"

CSCO installed NOTHING !!!!

"Products developed under the agreement were never integrated into Cisco's product line, according to Net2Phone. "

http://story.news.yahoo.com/ne...

LIAR !!

fatladysings 12/4/2012 | 10:47:10 PM
re: Cisco Eyeing Sonus Bond was convertible to equity
mrcasual 12/4/2012 | 10:48:08 PM
re: Cisco Eyeing Sonus
As such the stress tests where you send 64 byte packets (or whatever your favourite
small packet size is) continously for minutes/hours are meaningless.
____________________

Extended duration line rate tests reveal the worst case latency added by that network
element. How else does one make that emperical measurement?

+++++++++++++++++++++++++++++++++++++++++++++++++

Oops. Forgot about that one.

But then again an extended test with statistical traffic could find the same thing.

Most implementations I have seen do not usually have latency variations as a function of packet size. They usually exhibit throughput issues.
skeptic 12/4/2012 | 10:48:10 PM
re: Cisco Eyeing Sonus As such the stress tests where you send 64 byte packets (or whatever your favourite small packet size is) continously for minutes/hours are meaningless.
____________________

Extended duration line rate tests reveal the worst case latency added by that network element. How else does one make that emperical measurement?

=======================

If the purpose is measuring latency or testing
basic stability, its a
reasonable thing to do. But sometimes the
packet forwarding benchmarks seem to take
on a life of their own which is divorced from
reality.

From my point of view, I want the device to have
reasonable latency and be stable. The bar is
set at a certain level and vendors either
pass or fail. And pratically speaking I dont
know of a good reason for putting the bar at
100%.

Obsessing over some of these numbers (pick your
favorite packet size) can as often as not lead
in the long run to systems being misdesigned
to work well at a common test-point while it
falls to pieces everywhere else.
watchtower 12/4/2012 | 10:48:12 PM
re: Cisco Eyeing Sonus I also second the push to have LightMan stop the softsell on his web site (which consquently isn't very good).
watchtower 12/4/2012 | 10:48:12 PM
re: Cisco Eyeing Sonus optical_man, I second your opinion.

LightMan, you bring very little value to these conversations. It's evident you've never worked on or with these systems before.

Are you struggling for a career change?


>""Author: LightMan Number: 71
>Subject: Re: 2002 Core IP router requirements >Date: 3/12/2002 11:23:47 PM
>Don't know if I follow you. While I agree that >the info is high-level, I think it provides a >useful datapoint which vendors can use to >directionally assess their product fit. Granted, >it would be better if they delineated their >sources and questions, (probably could get it) >but my take is - I'll take what info I can.
>Not really a counterpoint - just a follow-on. >What is interesting is that it appears that they >are going to have trials to see how they meet >what they have interpreted as carriers needs.
>So given the vendors at large, who do you think >meets this criteria the best?""
>
>Lightman,
>OK, you're doing the softsell on your website, >www.thesoftwareprompt.com. Every one of your >messages has the same non committal tone to it.
>This last one is great! Right out of a marketing >slick:
>
>"..it provides a useful datapoint which vendors >can use to directionally assess their product fit"
>
>Please explain what the heck that means? A quick >question: What is a "datapoint" anyway? Is it a >point on a chart, or is it a silly word t make >one sound Perfeshenial.
>
>Coupled with your last comment " given the >vendors at large, who do you think meets this >criteria the best?"
>
>You seem to pose that question each and every >time, in varying forms. Something for us to think >about and visit your site to answer?
>Please stop. I learn a lot while I'm here, and >your softsell is getting on my nerves (take it >from a professional sales weenie).
>Thanks!!
rjmcmahon 12/4/2012 | 10:48:13 PM
re: Cisco Eyeing Sonus As such the stress tests where you send 64 byte packets (or whatever your favourite small packet size is) continously for minutes/hours are meaningless.
____________________

Extended duration line rate tests reveal the worst case latency added by that network element. How else does one make that emperical measurement?
mrcasual 12/4/2012 | 10:48:14 PM
re: Cisco Eyeing Sonus 100% line rates in practical situations are meaningless.
_________________

How does one build a packet network with deterministic latency characteristics if the
elements don't meet line rate?


I think the original comment was meant to convey the fact that in the real world you will NEVER see a line 100% utilized for any significant period of time. As such the stress tests where you send 64 byte packets (or whatever your favourite small packet size is) continously for minutes/hours are meaningless. Keeping up with a "normal", i.e. statistical distribution of traffic, is pretty easy.

The stream of packets tests are an interesting way to flush out "holes" in the datapath architecture. As an example 64 bytes might pass, but 65 might not. In the real world it is also possible to get reasonably sized bursts of back to back traffic. Stress tests are a convenient, although somewhat contrived, way to test the robustness of the datapath.

I don't know that I've ever seen a test where the packet size is swept over a range from say 32 to 128 bytes in one byte increments. I fully expect that if you did that you would see lots of performance holes in pretty much every datapath out there as internal bus/cell boundaries are crossed.
rjmcmahon 12/4/2012 | 10:48:15 PM
re: Cisco Eyeing Sonus 100% line rates in practical situations are meaningless.
_________________

How does one build a packet network with deterministic latency characteristics if the elements don't meet line rate?
skeptic 12/4/2012 | 10:48:30 PM
re: Cisco Eyeing Sonus Please explain. I am aware of the debate on packet reordering, and I've heard from both the Juniper rep, as well as the 9 Cisco guys I deal with on this. I'm thinking that it's a non issue (sortof like ATM is gtd, but packet is best effort, but I like packet better).
Do you have some insights that may enlighten me more?
-----------------------------------------

Ok. In my opinion, Juniper did something
(reordering) that should not be encouraged.
Everyone could take short-cuts in packet
forwarding or refuse (as Juniper has) to
fix a problem in their ASICs. But the network
ultimately suffers for their having done it.

Its possible that a particular person might
not see whats going on or that customers are
too lazy to see that their TCP sessions are
getting messed up, but it is a "bad thing"
for the internet to encourge vendors to
engage in behavior that if it becomes more
widespread is going to create problems.

And why should Juniper get credit for something
like 100% line rate on OC-48 and OC-192 as
"best forwarding". When most people know full
well that the 100% line rates in practical
situations are meaningless. I care that it
doesn't crash, but wire-speed at 100% line rate
on packets in benchmarks that
sometimes now less than the size of ATM cells means nothing to me.

So what I'm saying is that if I'm to give them
"credit" for best forwarding based on worthless
benchmarks that don't mean much in the real
world, why should I not give them a penalty
for breaking the rules and refusing to fix
a bug that they have known about for a long
time now.



Page 1 / 9   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE