1. Foundry Networks at JPMorgan 34th Annual
Technology Conference - Final
EHUD GELBLUM, ANALYST, JP MORGAN: Hello, thanks
everyone for coming again. Great you could join
us here late on a Wednesday. I appreciate it. My
name is Ehud Gelblum I'm the Data Networking Telecom
Equipment Wireline/Wireless analyst to JP Morgan,
and it's the last time in the -- last three days
that I'll be saying that. Very happy to have with
us Foundry Networks today, and with us the CFO Tim
Heffner and treasurer and Head of IR, Michael Iburg.
[We'll be] nice to Mike, he's thrown out his back
this morning and is on about a gazillion milligrams
of Motrin right now. If he says anything funny I
guess he's - that's under the influence. As in our
last presentation from Extreme, I'm very happy to
Foundry with us to talk to us about what the trends
that they're seeing enterprise market and the switching
market.
And without further ado, we're very happy to have
the two of them with us. So Tim, Mike, take it away.
MICHAEL IBURG, TREASURER, HEAD OF IR, FOUNDRY NETWORKS:
Thank you, Ehud. So as he mentioned, I did throw
out my back this morning, so I don't think I'm going
to say anything funny, I may grimace a lot so it
only hurts when I stand [or sit]. So, that's the
good news. So I've got a fairly brief presentation
here, its maybe 10 or 15 slides. Its - some [overview]
slides and then more product specific slides. And
for those of you who've seen us recently at a conference,
it's going to look fairly similar, but I think it's
a good sort of snapshot of kind of where we're at.
Okay, so before we begin, if we make anything that
sounds like anything other than a historical statement,
its probably a forward looking statement, if you
hear any of those, we always ask that they refer
to the most recently filed form -- documents with
the SEC in the form of a 10-K, a 10-Q, an 8-K. Those
documents have lots of good information about the
company, and we always ask that you refer to those
documents as you think about the company and the
things that we talk about here today.
In terms of the Foundry as a company in the data
networking space, we're an organization that's been
around for 10 years now, and we've focused primarily
on performance-based products or feature and high
value based products. So we've carved out a position
for Foundry in the market where we cater to high
end enterprise customers, high end universities,
hospitals, super computing sites, government agencies,
folks that are looking for a high performance product
for specific reasons or they're looking for high
features that are high value based products.
That's been the core market for Foundry and we
service the enterprise space, we service the Metro,
metro Ethernet solutions and the service provider
space. We service the content switching and Layer
4 through 7 market, as well as triple play on the
server provider side, and we've got some slides
around that.
So there's lots of areas where we touch the market.
But the core for the Found - the core of the company
has really been based on high performance, high
value, high feature solutions and that's how we
differentiate ourselves. Now at the same time, recognizing
that that's a market segment that's only so big,
we've been able to, over the last 12, 15, 18 months
try to take those features that exist in those high
end products and bring them into the middle of the
market.
So, we're primarily a company that targets the
medium and large enterprise customers. We get about
80% of our revenue from the enterprise space, get
the remaining 20% from the service provider market.
And within the enterprise space we're predominantly
selling Layer 3 enterprise LAN switches, which are
used in network upgrades. And then in addition to
that, we're selling a small component of Layer 4
through 7 products. S o what we're trying to do
in this market is really bring value, bring features,
bring things that other vendors either cant bring
or aren't bringing to the market, and deliver total
cost of ownership that is superior to what else
is available in the market.
In terms of the company, last quarter did about
114 million. Been profitable ever since we've been
public, generate a lot of cash, been in operations
about 10 years. We have about 10,000 customers,
the vast majority of those are enterprise customers,
probably more than 9000 of those are enterprise
customers. The company continues to get recognized
by the industry, because at the high end of the
market, we're often first to market with new technology.
So, way back in '97 we were the first vendor to
ever ship gigabit Ethernet. We were the first vendor
to ever ship a Layer 3 switch. We're the first vendor
to ever ship 10 gigabit Ethernet. So there's a pattern
of first to market. And by being first to market
we get a number of things, we get basically pricing
control. So we establish price in the market. We
capture all the early adopters; typically harvest
a higher gross margin, because there's not a lot
of competition when you're first to market.
So - and the other thing about playing at the high
end of the market is, that in the high end of the
market you're servicing a group of customers that
will often times upgrade their network more frequently
than your average rank and file enterprise customer.
In terms of the focus of the company, this is a
fairly detailed slide so I'm not going to go through
it exhaustively, but I just wanted to demonstrate
that the company has solutions that are specifically
targeted on the enterprise side at data networking
or infrastructure solutions that enable voice over
IP, enable wireless, enable the convergence of voice,
video and data on enterprise networks.
Power over Ethernet, we've talked about that on
a number of the most recent conference calls. Our
power over Ethernet port growth continued to be
very good. We set an all time record in the number
of power Ethernet ports shipped in the March quarter,
we announced that on the call I believe. Coming
around the horn, on the service provider side, and
I've got some slides to kind of walk through some
of this opportunity. We find ourselves playing in
areas where the service provider is looking for
more, either IP centric or Ethernet centric services.
So more and more of the MSOs and the traditional
service providers that are out there are looking
for solutions that leverage some of the advantages
of Ethernet. They're looking for solutions that
offer Ethernet routing and Ethernet peering. They're
looking at - although some of them are a little
more open than others, but 10 gigabit Ethernet as
a transport technology as opposed to OC-192, which
is a sonnet based technology for transport.
So, there's some areas where we see fairly significant
opportunities, we as it shows there, we play in
both the IP version 4 market and IP version 6 market
and we have some technical advantages in terms of
IP version 6, which I wont get into the details
there. And in areas where high density and high
throughput is a requirement, like IPTV or even commercial
data services, like Ethernet offerings to enterprise
customers. Areas like that, customers or service
provider customers are asking about things like
100 gigabit Ethernet, because further downstream
as technology and bandwidth continues to grow, aggregation
technologies that lead to 100 gig uplinks is something
that they're asking questions, and we're the only
vendor in the market that's shipping 100 gig capable
platform today.
So, although line cards aren't available today,
all of our MPLS router platforms today are architected
to accommodate 100 gigabits of in Ethernet per slot,
which means that we can sell them 10 gig line cards
today, but in two or three or four years when 100
gig line cards are available, they can simply buy
the line cards and put them into the chassis. They
don't need to do a forklift upgrade, they don't
need to have ability to swap out the chassis of
the platform, they simply buy new line cards and
upgrade to a higher speed technology. So that's
one of the advantages - one of the inherent advantages
that we have in the platform.
On the high performance computing side, not a huge
market but an area where we seem to do very well.
So we have a long list of customers that are doing
a lot of supercomputing where we sit either between
the super computer and the server farm or we provide
the transport mechanism but for large files, or
do large storage through the platform. In terms
of application management, we continue to round
out that offering, adding more features and functionality.
We announced some of that earlier this year, we'll
actually have some more of that coming out later
this summer.
And on the security side, all of our networking
products have a certain level of embedded security
baked into the platform. So within the hardware,
within the ASICs that power the boxes, there's a
certain amount of inherent [security] that we've
been able to put in, because many of our platforms
are actually powered by custom ASICs, and within
the customization of those ASICs we've been able
to embed certain security features that allow enterprise
customers, and occasionally service providers to
do better monitoring, understand bottlenecks, be
able to investigate or look into the traffic more
easily than in competitive platforms.
Just a couple slides on market trends. On the left
hand side, gigabit Ethernet is growing rapidly,
10 -- I'm sorry, 10/100 or what was commonly referred
to as fast Ethernet on the yellow line is actually
declining, has been for a couple of years. So the
transition in terms of technology is that 10 gigabit
Ethernet is growing as a core technology in the
enterprise. And a 10 gig becomes a predominant core
technology. Gigabit Ethernet becomes the primary
connection for all your servers, all your printers
and eventually all your desktops and laptops.
Does everybody need gigabit desktop? No, but most
vendors and Foundry included, no longer sell the
10/100 port. So when you buy wiring closet ports
from Foundry, they're all triple speed. 10/100 gig.
So whether you enable the gig or not, whether you're
- whether the adapter card on the piece of your
laptop can take advantage of a gig link or not,
that's up to t he enterprise, but in terms of port
deployments, the premium that you pay to put gig
to the desktop is not that great anymore. So most
people are moving in that direction.
In terms of power over Ethernet, the port shipments
are growing fairly dramatically, power over Ethernet
is the concept of delivering electricity down the
[inaudible] networking cable. So for an IT - for
voice over IP deployments, rather than having to
hook each IP telephone into a wall socket, you can
simply connect the data cable and the IP telephone
can be powered off the electricity coming on the
data line.
That's a great savings from enterprise perspective.
Even -- a better example is for wireless. Most wireless
access to points are deployed in the ceiling, and
if you have a fairly large facility and a large
square footage to accommodate, you my have 20, 30
or 40 access points in the ceiling. Well previously
you'd need to have a power source at each location,
we wanted to put an access point and that meant
bringing in an electrician and [inaudible] and conduit
through your rafters, but most IT organizations
could actually run data cable through the ceiling
themselves.
And the idea of being able to deliver power through
the data cable and being able to locate that anywhere
you want, and not having to have an outlet at any
location, means that you can put wireless access
points anywhere you'd like to put them. And if you
find that the reception isn't very good, you can
move it three feet over or five feet back or six
more feet to the right. And as long as you can stretch
the data cable there, you've got electricity there.
So most of the devices run [at] about 10 volts.
We pump 15, that's what's called a class 3 standard.
So we're seeing more and more of the ports move
in that direction. I think the last metric I saw
was about 20% of our total port shipments are power
over Ethernet ports. In terms of 10 gigabit Ethernet,
that market continues to grow. The port shipment
has continued to drive north as does the revenue,
the total market revenue. On the right hand side
you see the port pricing coming down. And we did
introduce a product at [M+I] a few weeks ago that
came in about $1250 per port, that doesn't include
the optical transceiver but --.
Pricing per 10 gigabit Ethernet is about five times
more expensive than gigabit Ethernet today. So for
five times the price you get 10 times the bandwidth.
So for the average enterprise customer who says,
I don't care about features, I don't care about
performance, I'm just cheap and I want the least
expensive solution I can find. Well today 10 gigabit
Ethernet offers the best value in terms of gigabit
per dollar spent. So 10 gig overtime will be come
a larger and larger, and a more predominant backbone
technology in the enterprise. We're beginning to
see that today, and I think some of the market research
numbers would convey that as well.
In terms of new products we introduced a handful
of them at M+I a couple of weeks ago. On the enterprise
side we introduced some new power over Ethernet
switches, and we also introduced a new family or
wireless products. And what we find is that as more
and more enterprise customers get experience with
voice over IP or get experience with wireless deployments,
they're getting more sophisticated in what they're
asking for.
So, we found that one of the concerns enterprise
customers have if with power over Ethernet, that's
a great concept, but if the power supply fails,
then those phones or that bank of phones dies. So
how do you keep the phones up and running? So these
stackables actually have redundant power supplies
for the power over Ethernet functionality, they're
hot-swappable, field upgradeable and self contained.
And they're the only ones like it in the industry.
Other data networking vendors will sell you a power
over - a redundant power supply module, which you
can stack on top of a box, but this is the only
self contained, Layer 3 LAN switch that has built
in redundant power for the power over Ethernet ports.
So that's one area where we try to - its not very
exciting, but its an area where we try to take advantage
of what we thought some of the vendors weren't delivering
into this market.
On the wireless side, we've introduced a series
of new wireless products. And one of the more interesting
ones is a mobility controller, which allows for
two things, it allows the enterprise customer to
manage all the wireless traffic and wireless access
centrally, in a centralized location. But also allows
them to basically control both the access to the
wireless network, and also the prioritization of
wireless users. Because one of the things that happens
is that the standards that were designed for laptops
conflict with the wireless standards.
So every time a wireless laptop tries to log onto
a wireless network, it reprioritizes itself as the
number one -- the most important laptop on the network.
And so network administrators have a tremendous
struggle trying to manage wireless traffic, because
everybody thinks they're the most important person
on the network. The mobility controller that we
introduced a few weeks ago actually allows enterprise
customers to set the priority of the user by IP
address or by some other metric.
So it gives them control above and beyond what
was available prior, and so it is something that
is not, again, not the most exciting or sexy product
on the market, but its something that addresses
another sort of gap where we saw an opportunity
where there's features or value where we could deliver
something that the other folks just aren't delivering.
For the service provider market we delivered a very
large box, the big black rectangle on the left hand
side. That's our XMR 32000. That [was] the market
as a demonstration of basically what we can do in
terms of port density, throughput and capacity.
As we've gotten deeper and deeper into the service
provider market, and spent more time talking to
both the traditional guys and the MSOs, we've learned
as we've worked through [IPTD] deployments or commercial
data service deployments, the port density and the
capacity at the edge is huge and the ability to
aggregate all that traffic and do it efficiently
is also important to service provider customers.
And to do it with a reasonable amount of rack space
because real estate is limited in a central office,
or in a POP or data center. And then to also do
it at a reasonable level of power consumption is
- some of the most important concerns that service
provider customers have today.
So the box was a good demonstration. The box will
ship for revenue a little bit later this year. But
it's a demonstration of the density both in 10 gig
port density and in gigabit Ethernet port density
that we can achieve, which far surpasses any of
the other platforms on the market. And then finally
on the Layer 4 through 7 slide we had some additional
announcements to sort of round out the portfolio.
Some of them were around a security feature set,
basically a software load onto a 4 through 7 platform.
And then some other applications that were designed
give the enterprise customer more control over the
traffic, more visibility into the flow of the traffic.
So, we get a lot of questions about the service
provider opportunity, where do we fit? Are we at
the aggregation layer or with the edge or with the
access edge? Or with the [edge] routing layer? And
in talking through this with a number of investors,
as well as with the sales folks who are actually
working on these deals, one of the things that we
sort of came to the conclusion on was basically
that its not an issue about where you fit in the
network. Its an issue about how can you solve a
problem for the service providers. And these are
4 areas where we see service providers building
out capacity.
Its areas where they have specific issues, either
economic issues or real estate or power consumption
issues or density issues. And there are other places
where they are building out capacity, but these
4 areas are areas where we've been able to engage
in a more [fulsome] way, to demonstrate what our
products bring to these markets.
So in terms of just basic consumer Internet, a
lot of these folks are - obviously this is perceived
as a commodity because everybody has Internet access,
everybody has broadband access. But one of the challenges
that service providers have is that broadband traffic
is still growing at a fairly high rate, but the
revenue stream isn't growing at the same rate. So,
a subscriber base is constantly demanding more bandwidth,
because they're consuming more bandwidth because
of the content that they're moving back and forth,
but yet they don't want to pay anymore money for
that.
So the economics of how do you deliver more bandwidth
in an economic fashion is a big concern for these
guys. And for Foundry, the opportunity here, I think
as its shown here is really around gig aggregation,
in terms of the layers, the level of aggregation
we can do and the densities we can achieve. The
cost advantage of 10 gigs of transport technology
relative to SONET or OC-192. And then on the Layer
4 through 7 side we have some fairly large deployments
in tier ones today. Basically sitting in front of
- sitting inside the data center, but sitting in
front of a server farm, basically feeding traffic
into a network; either sitting on front of a server
farm, a video server farm, farm, feeding traffic
into that network.
And so a lot of what we do today is trying to help
service providers accommodate just the demands of
residential broadband service. And so some of the
new products do a better job than the old products
do, and we're trying to get the new products more
commonly deployed in a lot of these accounts. In
terms of IPTV. It's a bandwidth intensive application.
There's two ways to do it, its either broadcast
video using multicast, which is not a big issue
because it can run okay on the existing technology.
But the other issue of video on demand using unicast
means you actually have to have an individual port
for each user, because you're basically establishing
a unicast connection with the individual. And that's
incredibly bandwidth intensive, and requires a huge
amount of port density. And so the platforms that
we have with the XMR and the MLX are platforms that
can achieve a level of density and aggregation that
cant be delivered by a lot of the other vendors.
And the other thing is that the price points make
it very economical, and the fact that the platform
is already architected, not just for 10 gigabit
Ethernet, which is the highest [inaudible] Ethernet
today, but its also architected for 40 gigabit Ethernet
if that's where the market goes or 100 gigabit Ethernet,
if it goes there as well. So, we've been working
on these products to make sure that they provide
sort of a stepping stone or a future [proofing]
methodology for the customers that we're selling
these products into.
Kind of moving over to voice over IP. The voice
traffic doesn't require a lot of bandwidth -- I
mean its not a bandwidth intensive application.
So its not a bandwidth play. This is probably not
one of our bigger opportunities, but the issue with
voice traffic is that the biggest issue here is
quality of service and the problem that most vendors
have is latency in the platform.
And one of the things that's been well documented
and well documented for a number of years, is that
architecturally Foundry's platforms have a lower
latency than every other platform in the market.
And there are some platforms in the market that
have a huge latency problem, where service providers
cant deploy specific platforms, because the latency
of the box would mean that the voice traffic, when
it got to the other end would be garbled or unintelligible.
So latency is an issue where, the box because of
its nature, actually slows down the traffic. And
its not a problem with data traffic, but when you
get the voice traffic it has a big impact on the
quality of service. So as service providers start
thinking about how to deploy voice services - voice
over IP services, and how they architect the network
that's going to give a quality of service similar
to the quality of service that's people experience
today with either POTS or PSTN, they're trying to
demonstrate or they're trying to look for solutions
that meet the technical requirements. And one of
the technical requirements is a low latency platform,
and that's one of the things that we offer today.
Finally, commercial data services, also known as
metro Ethernet, also known as enhanced Layer 2 [BPM]
services Layer 3 BPM services. Everybody's got a
different name for it, but it effectively comes
down to offering Ethernet services to enterprise
customers. The traditional guys are starting to
do it. The MSOs are asking the question can I build
out the capacity and basically take customers away
from the traditional guys.
And we're trying to demonstrate that in working
with some of these partners, some of those service
providers. We can demonstrate that you can build
out a certain level of capacity using our platforms,
and deliver a service level that the economics are
a lot better because of the cost of deployment relative
to a competitive solution.
And so the - some of the service providers out
there have already begun deploying Ethernet services.
Some of the feed back they're getting form enterprise
customers is that enterprise customers are now sort
of pushing back, asking for an Ethernet connection.
They're tired of T1s and T3s, they're tired of multiple
T1s or multiple T3s. they've got so much traffic
coming in and out of the network they want what's
commonly referred to as a fraction - fractional
gig interface.
So they want a fraction of a gigabit of Ethernet
as an Internet, from the service provider into the
enterprise. And we're working with folks trying
to help them understand if they haven't already
started, where they can go with this, what they
can build out, how the densities and the capacity
of our platforms would allow them to get to a better
economic model or better profit model. And so these
are the areas where we're trying to demonstrate
what our products can deliver.
So, from the service provider market in general,
we see -- in almost every instance, we see tier
ones building [on] capacity for service offerings
similar to what I've described here. A lot of them
are at different stages. The MSOs are in one level,
the traditional guys are a different level. They've
all got different priorities within their organization,
and we're working with each of them to figure out
kind of where we fit, and where our solutions can
be leveraged.
The company continues to drive products into the
enterprise space still [as] the bulk of our business.
And we have been successful with the products we've
introduced so far, and the enterprise growth will
-- the enterprise market will probably deliver the
bulk of the growth through the course of this year.
And as we've said on the most recent conference
call, we [think] there's a big opportunity in the
service provider market.
We think that's a good opportunity for us and we
can be successful there. But what we said on the
last call was that, we would expect that these new
products and these service provider products would
grow and add incrementally to the growth in the
second half of this year, but the real upside potential,
the upside opportunity is really 2007.
So with that, I will turn it over to the floor.
EHUD GELBLUM: Here's the floor.
MICHAEL IBURG: Here's the floor.
EHUD GELBLUM: All right. So I'll start off with
a question here and then we can --
MICHAEL IBURG: [inaudible].
EHUD GELBLUM: You sure.
MICHAEL IBURG: Yes, -- [reverse] my sentence.
EHUD GELBLUM: And then we'll see is there are any
questions from the room. You didn't talk about Japan
or geographically. A lot of people have had problems
with Japan, your Japan last quarter was strong.
What are you seeing there, what are the trends going
on there? [That is] not that big a part of your
business, but how do you - how does it impact -
what do you see going on in Japan trend wise? And
how do you think that plays out the rest of the
year? And why do you think you're seeing strength
right now, when everyone else seems to be seeing
weakness?
MICHAEL IBURG: Right so, we've done a couple of
things different in Japan in the last 12 months.
We've increased our direct sales force, added, its
not a large sales force, but we probably increased
it by about 50%. And since beginning of time, we've
always had one exclusive reseller who's really been
promoting our products in Japan. And beginning in
early '05, we basically brought on two additional
resellers who have different types of relationships
with the enterprise market.
And Foundry [inaudible] had a high concentration
of service provider revenue come out of Japan for
a number of years, two or three years ago. And it
was probably 75% service provider, 25% enterprise,
or maybe two-thirds, one third. And over the last
12 to 15 months, we've been trying to get back to
a more balance business in Japan. So the most recent
core to the March quarter, we're probably getting
closer to 50/50 where about half of our revenue
is coming from enterprise customers and half from
service providers.
So I think it's a combination of diversifying the
business away from a concentration of service providers,
it's about expanding the sales footprint and expanding
the channel relationships. So I think those are
the three things that we've done that have helped.
That said, I think we acknowledged on the conference
call that normal seasonality for Japan on the March
quarter is usually up. It's their fiscal Q4 and
we did in fact see an up quarter in Japan. And the
normal seasonality for the June quarter would be
down sequential, because it's their fiscal Q1. So
it's a small piece of the business, about 14 or
15% of our revenue and in prior years it's been
down slightly.
EHUD GELBLUM: Okay. Any questions from the room?
UNIDENTIFIED AUDIENCE MEMBER: [Could you] just
to follow up on that. Are you selling yen or in
dollars in [Japan].
MICHAEL IBURG: No we're selling them US dollars.
Yes?
UNIDENTIFIED AUDIENCE MEMBER: Can you discuss I
guess the deployment of some of the new products
that you've announced? And I guess ramp schedule
of some of the products maybe that you're shipping,
and whether those new products are addressing kind
of new markets that you really had low penetration
before or, just kind of upgrade some existing products
that you have?
MICHAEL IBURG: On the enterprise side or service
supplier side or both?
UNIDENTIFIED AUDIENCE MEMBER: Both.
MICHAEL IBURG: Okay. Well, that's a fairly broad
question. So - the new products, on the enterprise
side, the primary Layer 3 platforms are the Super
X and the RX and what they bring to the market is
they bring high performance, high feature set, a
high manageability into a broader enterprise market.
So Foundry's always been successful. And we've always
talked about five key verticals government, healthcare,
education, financial services and entertainment.
Those five verticals have historically been more
demanding on their network. They've been more sophisticated.
If you surveyed these guys, 85% of them would say
that they're smarter than the average bear. They'd
say that their IT staff is an upgrade from the average
IT staff, and they're looking for better solutions
or more sophisticated solutions.
If you go into some other verticals, like retail
and manufacturing, they'll tell you that they're
more sort of bread and butter, their more simple.
They're not looking for the high end stuff, they're
looking for more of the basic stuff. And the products
that we brought into the enterprise market are designed
to appeal to both those high end customers; from
a feature set, from a value, from a total cost of
ownership, from a performance level. At the same
time, appeal to a broader enterprise base just from
a price point perspective.
So the price points actually pull us into some
accounts, while the feature sets pull us into other
accounts. And a lot of it depends on what's important
to the customer, what's important to the end user?
So we found ourselves getting more success in certain
verticals where we haven't been as present in the
past, didn't have a stronger presence in the past.
And so that's been helpful, and we've talked oftentimes
about the number of new customers, and we've seen
a fairly steep increase in the number of new customers
over the last 12 months or four quarters.
And so we're confident that our competitive position
in the enterprise has improved because of the new
products. And it probably didn't hurt that some
of our competitors have been more challenged. On
the server provider side, the new products are being
deployed, we're shipping them for revenue. Oftentimes
at this stage the folks that are deploying them
in live networks are oftentimes either emerging
service providers or tier two, tier three type guys
because they wouldn't go through the exhaustive
testing that a tier one would go through.
But on the server provider side, we currently have
relationship with almost all the tier ones inside
the data center. So we've been selling Layer 3 equipment
and Layer 4 through 7 equipment inside the data
center to almost all the -- to a large number, I
wouldn't say all of them, but a large number of
the tier ones domestically and internationally.
So if you send a piece of email through a Comcast
or Time Warner, you'd find that that piece of email
would go through a Foundry Layer 3 switch inside
the beta center. And so - same thing would be like
SBC or Qwest would be another good example. Or AOL
or AT&T. The - inside the data center, you're
basically doing what's called back office services.
You're pulling video files off of a video server.
You're using a 4 through 7 deployment to basically
load balance that traffic, and then pushing it out
to - through a Layer 3 network out to the customer
facing network. And so, there's the data center
element which is sort of a self contained network,
and then you've got the customer [facing] wide area
which is the edge routing - well the access piece,
the edge routing piece, the core piece.
So we have a presence in almost all of - in a large
number of the service providers within the data
center. Because at the end of the day, we have a
reputation providing inexpensive reliable bandwidth.
And so for folks who see traffic growing and data
center traffic consuming the gear that they have
today, Foundry's been pulled into a lot of the accounts.
And so, we have a good reputation, a good relationship
with these folks. And what we're doing with the
new service provider products, namely the XMR and
the MLX is, we're leveraging the relationship we
already have.
And if you go a trade show and talk to our sales
reps they'll tell you we talk to these guys - we're
in these accounts talking to these guys on a regular
basis, but we're trying to leverage the relationship
that we have to parlay it into a lab, lab environment,
a trial environment or a pilot on the other side
of the fence, which is the customer facing commercial
network. So - and we're having some success with
that. I mean we haven't - we only made one announcement
and that was a service provider by the name of Limelight,
who [standardized] in our products for their network.
But I suspect second half of the year, we'll have
more customer announcements. Other questions.
UNIDENTIFIED AUDIENCE MEMBER: On the 10 gig market,
where's the early demand coming from? The second
question is, the cost of the largest barrier to
[wire] adoption and LRM is supposed to cut this
cost down a lot, are you going to get even better?
MICHAEL IBURG: Okay first - start over. First question
again?
UNIDENTIFIED AUDIENCE MEMBER: [inaudible question
- microphone inaccessible]
MICHAEL IBURG: A lot of the demand that we see
on the 10 gig side is enterprise network upgrades.
We have seen some government deployments, we had
seen some super company deployments, but today 80%
of our 10 gig - well maybe not -- maybe 70% of our
10 gig is basic enterprise upgrades, universities,
hospitals, fortune 500 type enterprises, folks like
that.
UNIDENTIFIED AUDIENCE MEMBER: [inaudible - microphone
inaccessible]
MICHAEL IBURG: Not any more. I mean at the end
of the day, a 5 to 1 relationship given that you
get 10 times the bandwidth is actually - its - that's
sort of the inflection point where we've seen it
in the past. So when we went from 10/100 to gigabit
Ethernet, when the price points got to about 5 to
1 that's when we sort of hit the ground running.
But the fact of the matter is, is that Foundry is
shipping its fifth generation of 10 gig products
and because of that, over the course of the subsequent
generations we've been able to drive cost out of
a platform.
So we're able to offer the product at a relatively
inexpensive price, although we still have reasonably
good gross margin, in some cases ahead of the corporate
average. And its because we've got the learning
curve or the experience of having done several generations
of products. A lot of our competitors are only on
their second or third generations, and there's an
800 pound gorilla that should be having a new generation
of 10 gig products on the market at some point.
And when they do that, then I suspect that they'll
be promoting 10 gig more aggressively. And the last
question, LRM.
UNIDENTIFIED AUDIENCE MEMBER: Yes.
MICHAEL IBURG: Don't know. Can't comment. Honestly,
I don't know. Yes?
UNIDENTIFIED AUDIENCE MEMBER: I guess 800 pound
gorilla [inaudible - microphone inaccessible]. Would
you imagine that the succession path recently is
going to stay over for that for years to come?
MICHAEL IBURG: Yes, you know we've always competed
against a large vendor in the space and we've never
been shy about that and we've always done well against
them. At the end of the day, we - most of the accounts
that we capture as a company come from Cisco. Because
at the end of the day, if somebody makes -- if somebody
was not happy with extreme or [inaudible], if they
went with an alternative to Cisco and it didn't
work out for them, odds are they're not going to
try an other alternative to Cisco, because the guy
who made the decision to go with an alternative
Cisco is no longer there. So the company is usually
going to go back to Cisco.
We do well finding folks that aren't necessarily
happy with the largest vendor in the market. Don't
feel they're getting the service or the support
that they'd like to have, don't feel like they're
as important as they'd like to be. Or in some sense
feel like they're being taken advantage of. So we
have the ability. And there's a large pool of people
that just aren't happy and we've been able to take
advantage of that.
I think that our large competitor will have new
products in the market at some point. We recognize
that. One of the things that we found in the past
is that, when they get the offering right and they
start pushing their installed base to upgrade to
a new solution, it oftentimes opens the door for
us. Because most customers today are at least, through
their own governance, would at least like to have
one competitive bid to make sure that they're keeping
the primary guy honest.
So we get oftentimes asked just to bid a network.
And I would say that we kicked out a lot of bids
knowing that we'll never get any business out of
it. But in about 15% of the situations where we
throw a bid into an account, whose told us I'm only
buying from my primary vendor but I'd like a bid
from you just to make sure that the pricings adequate,
and so we'll toss them a bid, and in about 15% of
the situations we win some business just because
of that.
So it's a foot in the door and if we're good at
it, we can parlay that into something more meaningful.
But we never shied away from competition. It's a
competitive market, always has been.
EHUD GELBLUM: Lets talk about that 800 pound gorilla
for a second. Their gross margins are in the - well,
we only see the blended gross margin, but they're
in the high 60s, granted there are a lot of pieces
in there, but thought wise is that where they compete
toe to toe with you their gross margins are higher
than yours, our '06 is 1.5%. How do you bridge the
gap between that? Is it just volume?
MICHAEL IBURG: So I don't - I don't know that they're
gross margin is that much higher than ours on the
comparable switching products. The - one of the
things that they can do and we cant do, is they
can take things like 10 gigabit Ethernet and if
you buy it in the 6500 platform, which is their
primary enterprise switch, they'll probably price
it at $5,000 or $6,000. I mean that's not too far
from - ours is a little bit lower than that, but
its comparable.
But in the 7600, which is very similar to the 6500,
the 10 gig pricing goes up to $7,000 or $8000, maybe
$9,000. Than in the 10,000 series product, the 10
gig interfaces are $15,000 or $20,000 and then the
12000 GSR family that same 10 gig interface is $40,000.
So we're selling it for $5000, or $4 to 5000 whether
its going into the enterprise space or the service
provider market. But they're taking a very comparable
technology that has a comparable transceiver and
component and in each platform they've got different
pricing based on the markets those platform goes
into.
And so if it's a core router, if it's an edge router,
if it's an enterprise box, if it's a metro router.
I mean they've just been able to establish different
price and at different levels. So I would imagine
that the gross margins in some of those examples
are fairly lucrative and more than we could ever
accomplish with where we're at.
EHUD GELBLUM: What's the magic of IP, versus Ethernet?
IP's more expensive for virtually the same thing.
MICHAEL IBURG: Right.
EHUD GELBLUM: Any other --?
MICHAEL IBURG: Any other questions before we leave?
Yes.
UNIDENTIFIED AUDIENCE MEMBER: [inaudible question
- microphone inaccessible]
MICHAEL IBURG: Okay well, in terms of - in terms
of federal government, I don't - we're not providing
an update. I mean our commentary on the call was
that good solid pipeline, we're engaged in a lot
of multiyear contracts. We thought we had a good
quarter in the March quarter. It was down sequential,
but it was only down a couple of million. We've
seen much more dramatic March quarters in the federal
space.
And in terms of forward looking, the comment that
we made on the call was that we said we thought
we'd have a very good year in 2006 relative to 2005.
But we didn't give any sort of calendarization around
that. So we didn't comment on the June quarter.
And in terms of Japan, I think it just comes down
to what we said earlier which was, we didn't see
the pause or the spending disruption that other
vendors have commented on.
So we had a good December quarter in Japan, we
had a good March quarter in Japan. And I think the
only commentary I really gave on the call was, we
simply referred to the fact that normal seasonality,
coming off of two strong quarters -- normal seasonality
for the June quarter would normally be down sequential.
So, yes its 14 or 15% of the business. Normal seasonality
on the enterprise side, both in Europe and domestic
enterprise would be up sequential and that's 50
- or 60% of the business. So we'll, most likely
- if normal seasonality played out, you'd have a
small geography going down, two large geographies
going up. So, we'll see how it plays out. Yes?
UNIDENTIFIED AUDIENCE MEMBER: [inaudible question
- microphone inaccessible] last quarter. I think
you're trying to rid of control, some were stock
option related. Can you just comment on where those
will go, directions?
TIM HEFFNER, CFO, FOUNDRY NETWORKS: I think I just
repeat what we said in the call - for those folks
that track the company, the expenses were up $5
million quarter-over-quarter. And we said that approximately
$3 million of those expenses were one time expenses.
However, in the second quarter those would be offset
by the three largest trade shows, the expenses incurred
to support the three largest trade shows of the
year. So we thought that in Q2 that the expenses
would be down. And then inQ3, because we don't have
those trade shows and we don't have those first
quarter one time that we had, we'll be down again.
So that was our commentary on the call.
MICHAEL IBURG: Yes?
UNIDENTIFIED AUDIENCE MEMBER: [inaudible question
- microphone inaccessible]
TIM HEFFNER: Well, that's the first time we heard
that question today.
MICHAEL IBURG: I'll turn that over to Tim.
TIM HEFFNER: Okay, hot potato. We do have a well
set policy for --
MICHAEL IBURG: For back dating - for how to back
date.
TIM HEFFNER: [inaudible].
MICHAEL IBURG: Okay, lets - we'll start from the
beginning so -
TIM HEFFNER: No, we do have a --
UNIDENTIFIED AUDIENCE MEMBER: [inaudible] Safe
Harbor, right?
TIM HEFFNER: Yes, that's right, no forward looking.
Well set policy for granting of stock options to
executives and employees. For those of you that
follow the company, you know Foundry is a conservative
company, and we're conservative in the application
of our accounting policies. This is fairly new thing
that we've seen. We haven't had a lot of time to
digest it, and so we don't have a lot of commentary
here today. Other than I can tell you that as of
now, we have not received any notifications from
any government agencies. Yes?
UNIDENTIFIED AUDIENCE MEMBER: How frequent did
you get together the records?
TIM HEFFNER: It's a very well documented - its
documented in all the minutes, because the board
pretty much oversees it - and or grants all the
options, so they'd be in the corporate minutes.
It'd be really easy to go after it and check it
our.
MICHAEL IBURG: And it think its -
TIM HEFFNER: All the officer filings are done for
the SEC on Form 4s.
MICHAEL IBURG: Yes, I think its -- those of you
who've looked into it obviously you can go to yahoo
finance and see all the grants because all the form
4s are filed, publicly available. They're also available
on the proxy as well. So, I don't think we're going
to comment any further on this. I think Tim kind
of offered kind of where we're at. And at that point,
I think we're out of time.
EHUD GELBLUM: One last thing before we go. Yes
you haven't given guidance, quarterly guidance in
about two years. What would make you change your
mind, and would you at any point consider giving
guidance again?
TIM HEFFNER: We do consider it, each and every
quarter we take a look at it. I think we've kind
of held back. We've been looking around to see what
others are doing. Everybody seems to be - they're
in, they're out --. I don't know what we're going
to do going forward, I don't have the answer today.
We do look at it each quarter.
And what would change us, change the dynamics?
I think a more stable environment. There's a lot
of uncertainty out there. I think a number of the
-- companies that have announced after Foundry made
people jittery, and its played out in a stock market,
pulling down all the networking stocks.
EHUD GELBLUM: Cool.
MICHAEL IBURG: Thank you.
EHUD GELBLUM: Thanks so much guys.
MICHAEL IBURG: Thank you, very much.
