[183160] in North American Network Operators' Group
Re: Drops in Core
daemon@ATHENA.MIT.EDU (Patrick W. Gilmore)
Sun Aug 16 08:01:01 2015
X-Original-To: nanog@nanog.org
From: "Patrick W. Gilmore" <patrick@ianai.net>
In-Reply-To: <20150815174129.GJ72293@-.true.nl>
Date: Sun, 16 Aug 2015 08:00:55 -0400
To: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
On Aug 15, 2015, at 1:41 PM, Job Snijders <job@instituut.net> wrote:
> On Sat, Aug 15, 2015 at 11:01:56PM +0530, Glen Kent wrote:
>> Is there a paper or a presentation that discusses the drops in the =
core?
>>=20
>> If i were to break the total path into three legs -- the first, =
middle
>> and the last, then are you saying that the probability of packet loss
>> is perhaps 1/3 in each leg (because the packet passes through
>> different IXes).
>=20
> It is unlikely packets pass through an IXP more then once.
=E2=80=9CUnlikely=E2=80=9D? That=E2=80=99s putting it mildly.
Unless someone is selling transit over an IX, I do not see how it can =
happen. And I would characterize transit over IXes far more =
pessimistically than =E2=80=9Cunlikely=E2=80=9D.
[Combining responses]
On Aug 15, 2015, at 1:21 PM, Owen DeLong <owen@delong.com> wrote:
>=20
> I would say that the probability of a packet drop at any particular =
peering
> point is less than the probability at one of the two edges.
>=20
> However, given that most packets are likely to traverse multiple =
peering
> points between the two edges, the probability of a packet drop along
> the way at one of the several peering points overall is roughly equal
> to the probability of a drop at one of the two edges.
I=E2=80=99m a little confused why most packets are =E2=80=9Clikely to =
traverse multiple peering points=E2=80=9D?
Most packets these days are sourced from one of three companies. (Which =
Owen should know well. :) At least one of those companies published =
stats saying the vast majority of packets are =E2=80=9Czero or one=E2=80=9D=
AS hop from the destination. I cannot imagine Google or Netflix being =
50% behind Akamai on that stat. Which clearly implies most packets =
traverse =E2=80=9Czero or one=E2=80=9D AS hop - i.e. one or zero peering =
points.
Finally, I would love to see data backing up the statement that packets =
are more likely to drop at one edge (assuming the destination?) than at =
a peering point.
--=20
TTFN,
patrick
