"It's nice to see that Vyatta isn't relying on open source's alleged good name along as it tries to carve out a prominent place in the networking realm. More young companies should have the guts to kick their rivals where it counts."

Unfortunately passe marketing tricks are about all Vyatta has going for them (and no I don't work for a gear vendor).

Their entire product line is based around commodity PC hardware. Performance is horrid compared to gear from any real vendor, where traffic is switched at a hardware level using dedicated ASICs.

The whitepaper on their site favorably compares the performance of a modern PC to an old, small Cisco router designed to route T1s (1.5mbit/s), something even a 486 could handle.

Apparently this means they "beat cisco."

Additonally there's really nothing their product offers that a competent administrator with a copy of FreeBSD couldn't brew up in an hour or two.

There is no way CPU based switching can handle the traffic of a Cisco 6500. There's a reason Real ISPs use Real Hardware, and not just off the shelf Dell boxes to route their traffic, and it's not because they don't have "super magic open source Vyatta software."

Yes, more young companies should stand up and whack their opponents on the conk, but part of the reason for not doing so might be that the young company is really looking to be bought by those same opponents.

Angry buyers are not likely buyers.

i had an ImageStream Linux-based router/firewall/VPN (with unlimited client licenses, all for under $6k) at my last gig, and it performed worlds better than the Cisco T3 router it was eventually replaced with (the Cisco was several times more expensive, and firewall and VPN were optional, and also expensive). my boss told me he wants the Linux box back (along with me, or someone at least as capable), but the new management at corporate HQ won't let him have it (or hire another BOFH-grade network type).

sorry if this offends the Cisco fanboys, but in my experience, Cisco devices are slow, expensive and poorly supported, and the support is hideously expensive too. if i had a question, i could call ImageStream and have a tech who knows the distro logged into my device within 15 minutes, checking things out. that's what i call support, and it only cost a couple-hundred dollars a year for 24/7/365 coverage. the ImageStream stayed up continuously for 4 years (i left around then). the Cisco failed after less than 12 months, and had to be repaired.

nothing personal, but most Cisco specialists don't know what they're missing. when people trot out those tired ASIC arguments, they neglect to mention that a recent networking speed record was set by a NetBSD box built on a generic PC platform.

finally, IOS (and all Cisco software, really) is still quite buggy and requires patching as much as Vista does. it may be a multi-billion-dollar global 800-pound networking gorilla, but Cisco has become bloated and complacent, and it's been that way for a long time. success has not been healthy for the company, regardless of profits. i'd rather have a responsive, agile vendor any day.

i haven't tried Vyatta's stuff, but maybe i should. if they're anything like ImageStream used to be, it could be a sweet product.

Maybe if Cisco would sell the 2821 closer to their real worth the comparison would have no validity. If they had compared it to a router designed to handle the same workload, the price difference would have been an order of magnitude higher.

Most Cisco gear seems to have that massive padding on top compared to other router makers, both software and ASIC brands, thanks to their 'no one ever got fired for buying Cisco' mentality. Don't mind us, though, if you're willing to get scammed that's entirely up to you.

I have used Vyatta and Cisco extensively, and I have probably completed one of the larger Vyatta deployments by a UK company (company WAN spanning 6 countries).

A few responses:

1) Vyatta vs Cisco, I am sure there is a point where the high end Ciscos out perform PC opensource based solutions, however at the commodity level this simply isn't the case. Bang for buck when you look at what you get with Vyatta feature and performance wise the equivalent Cisco is way way more expensive.

2) Vyatta vs Other Opensource, the argument is absolutely right, Vyatta does nothing that you can't get anywhere else on the opensource community, we actually use a Debian/iptables/imq solution elsewhere in our ISP business very successfully, and I would recommend if you are doing anything specialist you stick with the bespoke route. However Vyatta is extremely quick to deploy, give me a bare metal server, a Vyatta CD, and I will have an installed working Vyatta router with base config in 10 minutes, you simply can't do this with a custom build.

3) Cisco vs Opensource, long raging argument, we run our albeit small ISP all on Linux opensource routing equipment because its cheaper. Where Cisco wins though is support and the high end gear.

4) Support. Where Vyatta really wins is the support, if you decided go with Vyatta pay the subscription fee and get the support, its really superb.

Regards... Ben

@Michael H.

Sorry, But an Cisco 6513 isn't really a router only but a 13 slots chassis where you can install MAX 2 Supervisor 720's. I know we have 4 of them. You are comparing Apples and Peers.

They are high-end boxes not really the market where Vyatta is marketing for.

bt

"Where Cisco wins though is support and the high end gear.

4) Support. Where Vyatta really wins is the support"

WTF?