MMO servers...

[AreEss]

that was the worst of 3 runs. I see the jump in response time as well, at the telia.net peering point, but >200ms here.

Well, HE doesn't even really peer. They just pretend to. And do a piss-poor job of it. That kind of jump is a pretty good indicator that something is not right in the world.

I hit 'em slow.

Nice to see that Comcrap's still kept their incompetent netengs. Couldn't route tin-cans and string, those jackasses. (And still their customers scream how much better cablemodems are than DSL. Morons.)

Anyways, in the interests of (MOSTLY) full disclosure on my end, this is on my DSL, slightly loaded, at peak. The first two hops are edited, but the same physical router; DSL distant end. (This was performed from the router.)

traceroute to forums.amd.com (64.62.232.245), 64 hops max, 40 byte packets
 1  XXX.XXX.XXX (X.X.X.X)  12.933 ms  10.333 ms  10.210 ms
 2  XXX.XXX.XXX (X.X.X.X)  13.046 ms  12.934 ms  11.741 ms
 3  sl-gw4-roa-4-0-TS2.sprintlink.net (144.228.252.233)  23.115 ms  25.785 ms  22.861 ms
 4  sl-bb22-roa-2-1.sprintlink.net (144.232.17.205)  24.004 ms  23.980 ms  23.061 ms
 5  sl-bb22-chi-6-1.sprintlink.net (144.232.8.81)  26.064 ms  26.959 ms  27.002 ms
 6  sl-st21-chi-11-0.sprintlink.net (144.232.20.21)  26.800 ms  28.336 ms  27.341 ms
 7  144.232.8.98 (144.232.8.98)  27.747 ms  30.389 ms  26.008 ms
 8  hurricane-108836-chi-bb1.c.telia.net (213.248.84.46)  34.498 ms  25.964 ms  25.226 ms
 9  pos1-0.gsr12012.sjc.he.net (64.62.227.53)  94.589 ms  95.659 ms  98.252 ms
10  pos1-2.gsr12416.fmt.he.net (64.71.128.182)  96.239 ms  95.587 ms  94.463 ms
11  g1-0.gsr12008.fmt.fastservers.net (64.71.160.46)  95.332 ms  95.289 ms  95.070 ms
12  forums.amd.com (64.62.232.245)  95.534 ms  95.622 ms  95.821 ms

BTW, I'm sure you do have more exprience then I do, I actually see what you mean here, and agree that isn't a good thing. I don't have the experience with the equipment though (only ever worked with 2500 and 2600 series, and never in a real environment). Thanks for the insight

I'm a bit of an old hand with the BFR's, so HE's setup is that much more obviously crap to my eyes. Their 'state of the art' IPv6 is 7206VXR's with the old, weak NPE250. Meaning they can't even realistically get GigE line speed out of them. Not to say the 7206VXR isn't a good platform; it's one of the best routers ever created. The problem is the NPE. They should be running NPE350's or NSE1's. But I guess those cost too much - they're notoriously cheap.
The other fun is that people get mislead and confused by Cisco and HE. There's the 12000's, and the 12000XE's. The 12010XE is 50Gbit, but still obsolete. The 12012 did not come in the XE flavor, though, and taps out at 60Gbit on the backplane. That doesn't translate directly; BFR's are a funny beast like that. It has 12 slots, each slot taps out at 2.5Gbit/sec. That means two and a half GigE ports at linespeed. Oh, and 12 times 2.5Gbit is 30Gbit/sec out of the slots. Ignoring the fact that you do lose slots to your GRP.
Now, by comparison, the next hop is a 12416. Depending on configuration, the 12416 is 160Gbit or 320Gbit. Let's do the math here; 60Gbit goes into 320Gbit how many times? 5.3 times. It also pushes a lot more per slot. In other words, the logical arrangement places the 12416 as your demarc, distributing back in to Cat6500's with NatIOS. (Linecard thing; 12000-series only have optical GigE, and low port-per-linecard.) Or vice versa; a Cat6500 demarc (which saves you a lot of rackspace!) pumping into a 12416 core.

Frankly, it wouldn't surprise me to see ancient and many-decades-obsolete 7513's in their network as core routers. (We're talking stuff that doesn't even support GigE.)

 

[Xipher]

Yea, I had an idea about what you were saying, when you made the analogy of putting the cart in front of the horse.