On Thursday, August 25, 2011 02:46:27 PM Matt Moor wrote: > To my eye, the ASR1001 looks like a better option now > that it's available - it's an intel based platform (ala > RP2), albeit less powerful (Core Duo vs Xeon), with > something like the equivalent of an ESP5 inside it. > Talking to a few Cisco folk about it seriously (a few > months ago) more or less confirmed this. The 1002F > apparently had some market pressure driving it to > availability before RP2 was ready to go, and the 1001 > very much "feels" like an RP2 version of it (for those > of us in the Ethernet world, anyhoo). > > I left the company before we ended up deploying some, but > the due diligence had born out to that point.
Agree that the ASR1001 is certainly excellent on paper. We run 7201's as route reflectors here since 2008, and we're very happy that we're considering the ASR1001's as a potential replacement. It's just that the ASR1001 will only install 512,000 entries into the FIB, and we're not yet sure what a control-plane only router (route reflector role) will do when we exceed this maximum. In theory, there shouldn't be any issue since decisions are made in the RIB first. We have some 7201's in the edge that could have been changed to ASR1001's, but this 512,000 FIB entry limitation is a real PITA. The ASR1001 is on the right track, now we just need Cisco to fix that FIB issue, although something tells me they won't for fear of displacing the larger boxes in this series. Cheers, Mark.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
