You write as if a major carrier lets just anybody run an alien wave adjacent to all their other customers' circuits on two strands of fiber... Which is actually really rare. You need to have a high level of trust and confidence in the abilities of the optical engineering on both sides. It is true that if you buy a DWDM wavelength provisioned between a set of chassis on both ends (Ciena, Adva, whatever) it will be indistinguishable *to you* from a L2 ethernet circuit. Whatever device you put on one end with a single MAC address, or a set of MAC addresses, will show up on the other end. Usually provisioned with 1310nm LX optics on both ends for the greatest cost effectiveness.
It is entirely possible that an inter-city transport provider could put you on either a 10 Gbps circuit inside a 100 GbeE PTP that they run, or on a legacy 10 Gbps wavelength DWDM chassis system (Ciena 4200 anyone?). You should not be able to tell the difference as the end user. On Wed, Dec 20, 2017 at 7:48 AM, Seth Mattinen <se...@rollernet.us> wrote: > On 12/19/17 10:00 AM, Carl Peterson wrote: > >> We are about to sign a contract on a 10G wave and I'm trying to figure >> out what the wan phys should be and whats the difference. I'm assuming we >> would want lan phys. >> >> Does this make sense and would we just use normal 10G optics on each end? >> > > > If you're getting a true wave it's neither LAN nor WAN PHY. You will get a > wavelength, literally. I recently ordered some 10G waves which were > specified as requiring 1310nm, 1330nm, and "1550nm reaching 80km" optics. > > If it's LAN PHY you're getting plain old 10GbE Ethernet like with a > switch. If it's WAN PHY you're getting SONET and the carrier will probably > use OC-192c/STM-64 to carry it. > > ~Seth >
