More on the NX64000. PSInet is to use it. See press release below.
I recall the Monterey claims when they were first released. Their WaRP protocol elicited similar excitement. What that turned out to be was a take off on multiprotocol label switching, or MPLS-like, at the lambda level. And since lambdas are now almost synonamous with the outer optical carrier limit, or OC-x.. OC-192 in this case, then the port speed defines the speed at which they are able to "route." The trick then becomes being able to concatenate or define how streams are organized (framed to some SONET or other format) electronically, and then at the optical port level, at ever higher speeds. OC-768 is state of the art. Faster designs are already on the books.
Some not so obivous by products here have to do with the ability of optical amplifiers (erbium doped fiber amps) to perform at these speeds due to dispersion artifacts vis a vis the implications of spacing them on the actual routes. Distances tend to get smaller as speeds increase, in other words. These parameters would be felt in the transoceanic space, where distances are pretty much fixed, already, on those cables which have already been placed.
MPLS is the IETF's answer to Cisco's Tag Switching protocol. In either case, as I suspect is the case here, this is a Layer 2 process under the direction of Layer 3 addressing pointers. This is sounding similar. Very powerful stuff, though, no matter how you want to cut it. Frank
cbs.marketwatch.com |
