11–20 of 53 posts
Wonder what location he tunneled from / to to get around Verizon's bottleneck.
Hi! OP here. I'm in NYC. I tunneled to Washington D.C. and that seemed to do it. I figured I'd try a location as close as possible first.
Also would like to note we pay for (and can get) 50 Mbps down. Netflix comes in at ~560 kbps and just stays there.
I think this issue may be more subtle than we want it to be. I'm no fan of monopolistic ISPs, but it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. You pay per byte sent, just like how the sender of a letter pays postage. This has always been the case, and as long as Verizon charges Level 3 the same amount per…
Up until very recently the only time you paid for peering was when you were actually buying transit. Level 3 is definitely on the right side of this argument, not only logically - but also from the perspective of precedence. Anybody interested in learning how the internet actually works should read "The Internet Peering Playbook".
Earlier quoted context omitted.
Hi! OP here. I'm in NYC. I tunneled to Washington D.C. and that seemed to do it. I figured I'd try a location as close as possible first.
Tried the same thing on TWC from NYC with no difference in speed. Will experiment with other setups. Also would like to note we pay for (and can get) 50 Mbps down. Netflix comes in at ~560 kbps and just stays there.
I think this issue may be more subtle than we want it to be. I'm no fan of monopolistic ISPs, but it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. You pay per byte sent, just like how the sender of a letter pays postage. This has always been the case, and as long as Verizon charges Level 3 the same amount per…
> ...it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. Up until very recently the only time you paid for peering was when you were actually buying transit. Level 3 is definitely on the right side of this argument, not only logically - but also from the perspective of precedence. Anybody interested in learning…
I think this issue may be more subtle than we want it to be. I'm no fan of monopolistic ISPs, but it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. You pay per byte sent, just like how the sender of a letter pays postage. This has always been the case, and as long as Verizon charges Level 3 the same amount per…
Such an arrangement with a consumer ISP is so preposterous I have to assume that Level3 and Verizon never expected flows to be symmetric.
Level3...isn't paying for that difference.
Right, because Verizon's customers pay for the difference.
Earlier quoted context omitted.
> ...it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. Up until very recently the only time you paid for peering was when you were actually buying transit. Level 3 is definitely on the right side of this argument, not only logically - but also from the perspective of precedence. Anybody interested in learning…
Hmm, I guess I don't understand why sending packets to an end user over an ISP is not "transit", but I'll go ahead and admit I know very little about this subject and am probably wrong.
I don't know enough about the issue to say whether that's true or not, but that seems to be the argument they're making.
Earlier quoted context omitted.
> ...it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. Up until very recently the only time you paid for peering was when you were actually buying transit. Level 3 is definitely on the right side of this argument, not only logically - but also from the perspective of precedence. Anybody interested in learning…
Hmm, I guess I don't understand why sending packets to an end user over an ISP is not "transit", but I'll go ahead and admit I know very little about this subject and am probably wrong.
Earlier quoted context omitted.
> ...it is true that when you have a peering point with unbalanced traffic flows, the side sending more traffic is supposed to pay the side receiving more traffic. Up until very recently the only time you paid for peering was when you were actually buying transit. Level 3 is definitely on the right side of this argument, not only logically - but also from the perspective of precedence. Anybody interested in learning…
Hmm, I guess I don't understand why sending packets to an end user over an ISP is not "transit", but I'll go ahead and admit I know very little about this subject and am probably wrong.
Earlier quoted context omitted.
Hmm, I guess I don't understand why sending packets to an end user over an ISP is not "transit", but I'll go ahead and admit I know very little about this subject and am probably wrong.
If it were considered "transit" in the same sense as delivering packets from one network to another across large distances, then consumer Internet access should be free because it's always biased in favor of receiving rather than transmitting bytes. Verizon should even be paying its customers to accept the extra bits they get.
But yeah, I clearly don't know what I'm talking about at this point. Sorry.