Market Leader Programs
5G Transport - A 2023 Heavy Reading Survey
2023 Open RAN Operator Survey
Coherent Optics at 100G, 400G, and Beyond
Open RAN Platforms and Architectures Operator Survey
Cloud Native 5G Core Operator Survey
Bridging the Digital Divide
5G Network Slicing Operator Survey
Open, Automated & Programmable Transport
The Journey to Cloud Native
JJ
and those who stay HAVE A LOT OF MONEY.
If a Procket falls in a forest, does it make a sound?
Simply, they failed to get any RBOCs to invest in them.
They had people from Qwest, MFN, UUNET, and other companise invest, but noone from the RBOCs.
Given the failure of the 2nd tier ISPs, and their eventual replacement by RBOCs and PTTs, I think this is the main reason Procket will fail.
Something like 20 ISP Engineers and Executives invested in Procket's last round, contributing something like 5% of the capital that the company raised.
-> Edge MPLS stuff
-> Private route tables (VPN) and lots of them
-> Multicast
-> Diffserv
-> NAT
-> More tunnels (types, numbers and combinations) than you would care for
What happens if those apparently simpler Edge ASICs don't work ? Does a company move along to building an Enterprise/access box ?
And what about software ? An edge box isn't merely a slower core box, as the article seems to suggest. Have we hired OSPF chaps ?
And the Edge market is a piece of work by itself. Apart from Cisco and Juniper, you have the mid 90s startups that are gaining traction (Yago->Cabletron->Riverstone, Redstone->Siemens->Unisphere) plus other startups that seemed to be getting somewhere before losing steam (Gotham, Celox .. the list is endless). Sure, there could be some space for another player, but at some point you're going to have a fairly large set of moderately stable solutions. And what happens then ? Margins get to be more reasonable and how do we justify $200m funding rounds then ?
No one on this board seems to have drawn the analogy yet, but does this remind anyone of Caspian Networks ? You can't deny them their hubris, but if I were to ask a bunch of VCs for $200m, I wouldn't say we're building a better edge router. I'd say we're building the box to replace all boxes (ala Caspian).
Actually, with all that money, they could slow things down and be like Pluris, the edge box vendor that just doesn't go away.
What you gain by going to the edge is that
you can buy lots of the necessary pieces off
the shelf. Also, the central switch in the
box at edge-type capacities requires a whole
lot less R&D. You can buy most of the ASICs
rather than scratch-build them. But then again,
so can everyone else.
However, the scary thing about edge products
vs. core products is that you have to do "more"
of everything. You have to do every physical
media and routing protocol that anyone can think
of. In total, it adds up to a whole lot more
software work than a core router and a whole
lot more hardware/hardware-integration projects
going on.
And its not like it was a couple years ago.
Riverstone and others are getting well-entrenched
in that space and while their products are not
perfect, they are going to fight like crazy to
protect their piece of the market.
Procket is in a tough spot though. Their product
isn't big enough to fight in the war for the
next-generation core router. And going
head-to-head with juniper at the midrange core
would be very tough. Unfortunately, if they
go toward edge routers, they are coming into
the game very late.
It's never easy to build ASICs. And word was that Procket was doing full custom stuff, which of course makes it quite likely that there'll be early problems. I would not write them off. Probably they're winning time (and revenue) by leveraging their SW in a market niche that allows them to get to market faster with some off-the-shelf stuff, while they also debug their custom stuff more leisurely. Why bet the farm on 40G now when you have other options? And when and if the core market turns around in time, they'll be as well positioned as a start-up can be.
Does look like a smart move to me.
It's a very hostile environment for *every* start up out there right now. Those who claim they've got it made either suffer a severe case of tunnel vision, or they are in denial. Good start-ups these days are those that leverage what they already have as opportunistically as possible, re-align themselves strategically, and all the while do not entirely give up on the original plan.
Procket missed the boat. Companies aren't buying startups for superstar talent anymore. Chambers himself said Cisco will be acquiring companies with REVENUE.
It'll take another $20M-$30M+ for Procket to get an edge router to market and probably a bridge loan to get to an IPO. Talk about dilution ...