Our dashboard for world IPv6 day is now officially open at http://v6day.ripe.net/ It leads to loads of statistical and diagnostic information. We are still tweaking it a little and adding things. Enjoy Robert
On Jun 7, 2011, at 11:02 AM, Robert Kisteleki wrote:
Our dashboard for world IPv6 day is now officially open at
It leads to loads of statistical and diagnostic information. We are still tweaking it a little and adding things.
Enjoy
Robert
Looks like somebody is doing test runs across ams-ix or did some private peering flap ? Anybody in the EIX community that sees the same or who can shed some light on this ? Marco (just a curious citizen)
On Tue, Jun 7, 2011 at 10:08, Marco Hogewoning <marcoh@marcoh.net> wrote:
Looks like somebody is doing test runs across ams-ix or did some private peering flap ? Anybody in the EIX community that sees the same or who can shed some light on this ?
According to a source on IRC it was a measurement error (confirmed by an AMS-IX statement). -dominik
On 7 Jun 2011 (24), at 10:08 AM, Marco Hogewoning wrote:
Looks like somebody is doing test runs across ams-ix or did some private peering flap ? Anybody in the EIX community that sees the same or who can shed some light on this ?
It was test traffic and should not have been counted. We are working to subtract it from the real traffic. -- Arien
Excellent! Any idea what happened on AMS-IX? Did someone started testing one day earlier? -- Tassos Robert Kisteleki wrote on 07/06/2011 10:02:
Our dashboard for world IPv6 day is now officially open at
It leads to loads of statistical and diagnostic information. We are still tweaking it a little and adding things.
Enjoy
Robert
Hi, cool dashboard! Thanks. I was wondering if you have considered (or could consider) also adding tracepath6 / scamper pmtu discovery probes to the measurements as well in some fashion. A non-trivial exercise, I'm sure. Would be quite useful, I imagine. Cheers, Martin On Tue, Jun 7, 2011 at 3:02 AM, Robert Kisteleki <robert@ripe.net> wrote:
Our dashboard for world IPv6 day is now officially open at
It leads to loads of statistical and diagnostic information. We are still tweaking it a little and adding things.
Enjoy
Robert
Hi, On 2011.06.07. 16:52, Martin Millnert wrote:
Hi,
cool dashboard! Thanks.
I was wondering if you have considered (or could consider) also adding tracepath6 / scamper pmtu discovery probes to the measurements as well in some fashion. A non-trivial exercise, I'm sure.
Would be quite useful, I imagine.
Cheers, Martin
Considering the fact that we are at T-8h or so, I think we'll look into implementing this next time :-) Regards, Robert
Robert, On Tue, Jun 7, 2011 at 11:39 AM, Robert Kisteleki <robert@ripe.net> wrote:
Hi,
On 2011.06.07. 16:52, Martin Millnert wrote:
Hi,
cool dashboard! Thanks.
I was wondering if you have considered (or could consider) also adding tracepath6 / scamper pmtu discovery probes to the measurements as well in some fashion. A non-trivial exercise, I'm sure.
Would be quite useful, I imagine.
Cheers, Martin
Considering the fact that we are at T-8h or so, I think we'll look into implementing this next time :-)
Fair enough :) I was just wondering if you had it already just waiting to push the big launch button. So now that the measurements are ongoing and I see like ~everyone is using akamai proxies, and the overall relative RTT is ~0.8, I have a request which is pretty reasonable i think (which should be easy enough to do in post-processing): Group a summarized relative RTT stats graph by those who use Akamai for this and those who don't, to effectively benchmark their service/peering/ etc. (Obviously, since so many sites use the same single provider for this day, that single provider influences stats a lot). Best, Martin
So now that the measurements are ongoing and I see like ~everyone is using akamai proxies, and the overall relative RTT is ~0.8, I have a request which is pretty reasonable i think (which should be easy enough to do in post-processing):
Group a summarized relative RTT stats graph by those who use Akamai for this and those who don't, to effectively benchmark their service/peering/ etc. (Obviously, since so many sites use the same single provider for this day, that single provider influences stats a lot).
We have the data and plan to do a number of post-op analyses. We'll take your idea on board as inspiration! Cheers, Robert
Robert, great! On Wed, Jun 8, 2011 at 2:01 AM, Robert Kisteleki <robert@ripe.net> wrote:
So now that the measurements are ongoing and I see like ~everyone is using akamai proxies, and the overall relative RTT is ~0.8, I have a request which is pretty reasonable i think (which should be easy enough to do in post-processing):
Group a summarized relative RTT stats graph by those who use Akamai for this and those who don't, to effectively benchmark their service/peering/ etc. (Obviously, since so many sites use the same single provider for this day, that single provider influences stats a lot).
We have the data and plan to do a number of post-op analyses. We'll take your idea on board as inspiration!
I realize there are two kinds of usage going on, which I'm sure you do too: 1) akadns for geolocation (I presume), 2) actually serving pages from akamai Would be interesting to hear more from them. Cheers, Martin
participants (6)
-
Arien Vijn -
Dominik Bay -
Marco Hogewoning -
Martin Millnert -
Robert Kisteleki -
Tassos Chatzithomaoglou