re: Li Quits ProcketLet's face it. Tony was a figure head at Procket. For much of the time he was managing the driver group - it wasn't as if he was crunching heavy-duty BGP code.
Many operators respect Tony and he is definitely valuable from that perspective, I guess the Investors believe they get that with Roland; though not sure it is an apples to apples swap.
If I was a procket fan I'd be more worried about some of the teething problems WRT the technology than anything Tony does or doesn't do. It is great to build a fast machine, but it has to fast under all conditions - Procket has some interesting design issues here. Also for the largest carriers (perhaps a small number admittedly) the lack of a cluster solution is a problem.
So if Procket fails from here, it is not because Tony left, it is because being another Juniper is ultimately not that compelling - being another Juniper with technical clitches and no clustering solution is even less so. I'd be more worried about that than anything else.....
re: Li Quits ProcketHello, I be John Chambers personal assistant, ya may remember me from such strategic acquisitions as Cerent, Stratacom, an' recently Hammerhead networks.
First o' all, I'd like ta send uh shout out ta all muh ma fukin peeps on Sand Hill cruize. Fo' Shizzle ma nizzle homes. Keepin' it real up in Menlo Park. Fo evuh.
Second, I gots some exciting news fo' all da tasty little peeps out dere. We iz increasing our investment in Procket, ta 99.9% o' Procket. For me ta poop on. We iz purchasing all o' procket except da following assets: a) da stack o' 38,000 unused "Inspected By Tony Li" stickers b) da procket fish. We don' need fish, we's iz not uh fish company, fish iz not our culture. Repeat 5 times, rinse an' lather. As part o' dis here deal we's will also port da BSD procket layer ta an internet enabled toaster over near ya.
All yo' routing update iz belong ta us!
Roland thought he could just slip away from us without paying up da office NFL pool. Tisk Tisk, silly Roland, since ya stiffed us on dat $50 ya will git dat crappy office in da basement o' Building 10 where da air conditioning don' werk, an' da wall vibrates whenever someone flushes da toilet. You will return da stapler as well! Cough up da dough dammit.
All yo' BGP iz converge on Tasman.
Embrace yo' new networking overlords! Bow down 'bfoe thy diety IOS!
re: Li Quits Procket"From what I know of Tony ( mostly nanog meetings) he is a straight shooter and very approachable. An engineer's engineer."
They rarely (and never in isolation) build a successful & profitable company....
Maybe someone should let the guys who are half as good as he get on with what they are paid to provide and if all is said and done and everything is as true as the rumors suggest Procket should walk this mile of discontent and still win.
What is that saying again? In the land of the blind the one eyed man is King? Maybe its time one of the blind poked the king in the eye..........
re: Li Quits Procket<quote> If you read the press release, it clearly states that this is not a 1 or 2 box deal. More like 12. AARNET is a very large backbone, and it is a significant win for Procket </quote>
re: Li Quits ProcketWell let's see - 12 8812's = 480GB per chassis x 12 = 5760GB total.
With a 12416, 16 slots @ 10GB minus 2 for RP's = 14 x 10GB = 140GB per chassis. So 5760 / 140 = 41 12416 chassis to equal the capacity of 12 8812's.
Hang on, why do you assume fully loaded 1/2 rack boxen?
The Procket story mentions "8000-series" routers only, no type, no capacity.
If you think a university needs and can afford 12 fully loaded core routers... we're talking a carrier's core network here, hundreds of companies and millions of consumers worth of traffic requiring that many boxes. An Australian university has no use for that amount of gear.
Many operators respect Tony and he is definitely valuable from that perspective, I guess the Investors believe they get that with Roland; though not sure it is an apples to apples swap.
If I was a procket fan I'd be more worried about some of the teething problems WRT the technology than anything Tony does or doesn't do. It is great to build a fast machine, but it has to fast under all conditions - Procket has some interesting design issues here. Also for the largest carriers (perhaps a small number admittedly) the lack of a cluster solution is a problem.
So if Procket fails from here, it is not because Tony left, it is because being another Juniper is ultimately not that compelling - being another Juniper with technical clitches and no clustering solution is even less so. I'd be more worried about that than anything else.....