[147550] in North American Network Operators' Group
RE: Multiple ISP Load Balancing
daemon@ATHENA.MIT.EDU (Drew Weaver)
Wed Dec 14 14:45:23 2011
From: Drew Weaver <drew.weaver@thenap.com>
To: 'Christopher Morrow' <morrowc.lists@gmail.com>
Date: Wed, 14 Dec 2011 14:44:27 -0500
In-Reply-To: <CAL9jLaYf13sWz=20JrufO=W+qfHXeFBEYoUesuhrONwzgdXcZQ@mail.gmail.com>
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
>seems the feeling is that if you have multiple full feeds and need to load=
share, you really don't want (in most cases) ispa=3D500mbps + ispb=3D500mbp=
s.
>
>
>you really want destinationA to be reached across the 'best path'
>(best ... in some form, distance? packetdrop%? jitter? cost?) you'll most=
likely have to tweak things in order to achieve what you want since only d=
istance is really used in the stock bgp calculation (distance by as-hops, p=
resuming you don't listen to closely to med from your providers)
Yes, but performance from your network to $destination_AS via $ISPx can be =
variable and how do you know when it changes before someone starts complain=
ing?
There are traditionally two pieces involved with optimization.
1) "Cost" (Commitment/oversubscribe management and monitoring)
2) "Performance"
Usually "cost control" is #1 so systems like that are configured so that as=
long as the traffic isn't breaking your commits or filling your pipes they=
will then optimize X number of your top prefixes for performance (based on=
what the system can see).
The performance aspect is generally just sending basic probes in all direct=
ions towards a destination host and seeing which ones reply the fastest.
Although obviously this only impacts traffic outbound from your AS.
-Drew
