Re: Transit SLA
"Jere Retzer" <[email protected]> Wed, 16 Jul 2003 08:58:07 -0700
| Newsgroups | gmane.org.operators.rep |
|---|---|
| Message-ID | <[email protected]> |
This is a MIME message. If you are reading this text, you may want to consider changing to a mail reader or gateway that understands how to properly handle MIME multipart messages. --=_035D2437.7714715F Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable From what I've read/heard elsewhere it seems to me that traditional = "peering" exchanges would not have a SLA but that this is appropriate for = "transit" exchanges Anthony de Boer wrote: >>Multihomed through a single local loop? For redundancy, I've always leaned toward using upstreams who share no infrastructure and are barely even talking to each other. Definitely a link to an exchange is a good thing, but a second loop going elsewhere makes the backhoe work harder. In our case, we're working on additional metro locations connected using = GE over dark fiber. This allows us to come up with a fully redundant, = multi-homed configuration if suppliers and customers both connect to two = locations. Customer connects to location A and C, for example and = suppliers connect in any two of the available locations, say A and B. We are, for additional redundancy setting these up as dual parallel = networks. Continuing the above example, customer connects to network one = at location A and network two at location C. Suppliers additionally = connect to both networks, one at each location. Given this arrangement, = any single local loop, or network can go down and the customer will still = be served and may in addition use however many different suppliers they = desire with the same two local loops.=20 I'm curious if others on this list are using/experimenting with similar = approaches and if so, are you using some sort of ethernet ring protocol? = Is this sort of feature worth the added equipment cost, given that the = available solutions are largely proprietary and require more expensive = switches? Jere --=_035D2437.7714715F Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Description: HTML <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1"= > <META content=3D"MSHTML 5.50.4727.700" name=3DGENERATOR></HEAD> <BODY style=3D"MARGIN-TOP: 2px; FONT: 12pt Verdana; MARGIN-LEFT: 2px"> <DIV>From what I've read/heard elsewhere it seems to me that traditional=20= "peering" exchanges would not have a SLA but that this is appropriate = for=20 "transit" exchanges<BR><BR>Anthony de Boer wrote:<BR><BR>>>Multihomed= =20 through a single local loop? For redundancy, I've always<BR>leaned = toward=20 using upstreams who share no infrastructure and are<BR>barely even talking = to=20 each other. Definitely a link to an exchange<BR>is a good thing, but = a=20 second loop going elsewhere makes the backhoe<BR>work harder.<BR></DIV> <DIV>In our case, we're working on additional metro locations connected = using GE=20 over dark fiber. This allows us to come up with a fully redundant, = multi-homed=20 configuration if suppliers and customers both connect to two = locations.=20 Customer connects to location A and C, for example and suppliers connect=20= in any two of the available locations, say A and B.</DIV> <DIV> </DIV> <DIV>We are, for additional redundancy setting these up as dual parallel=20= networks. Continuing the above example, customer connects to network one = at=20 location A and network two at location C. Suppliers additionally = connect to=20 both networks, one at each location. Given this arrangement, any single = local=20 loop, or network can go down and the customer will still be served and may = in=20 addition use however many different suppliers they desire with the same = two=20 local loops. </DIV> <DIV> </DIV> <DIV>I'm curious if others on this list are using/experimenting with = similar=20 approaches and if so, are you using some sort of ethernet ring protocol? = Is this=20 sort of feature worth the added equipment cost, given that the available=20= solutions are largely proprietary and require more expensive switches?<BR><= /DIV> <DIV> </DIV> <DIV>Jere</DIV></BODY></HTML> --=_035D2437.7714715F--