Hi, I have compiled an in-depth analysis of peeringdb data. You can find a full description of the method, scripts, data and the main results on github:
https://github.com/mwichtlh/address-policy-wg/ Some interesting takeaways: * Roughly 83% of all IXPs would theoretically fit into a /25. This already includes 100% overprovisioning, i.e., 2xconnected ASes/IXP. At the same time, 74% of all peering LANs are /24s. Consequently, the default policy of assigning /24s has created large amounts of unused space. * Already today, more than 10% of all peering LANs are smaller or equal a /25. Having small peering LANs is not entirely unusual. * Large IXPs requiring a /23 or larger are rare (<3%). Thus, lowering the upper bound for assignments to /23 will not save large amounts of space. Conclusions: I back the proposal except for the limitation to a /23. I propose having a /21 as an upper limit with thorough checks by RIPE. Regards, Matthias On Wed, 2019-06-05 at 03:47 -0700, Randy Bush wrote: > > > Hmm.. why shouldn't defunct IXPs not be taken in consideration > > > though? > > > > Because they will have handed back their address space. > > what are you trying to measure? the space utilization of current > operating exchanges, or the distribution of request sizes? > > randy > -- Dr.-Ing. Matthias Wichtlhuber Researcher ------------------------------ DE-CIX Management GmbH Lindleystr. 12, 60314 Frankfurt (Germany) phone: +49 69 1730902 mobile: +49 171 3836036 fax: +49 69 4056 2716 e-mail: [email protected] web: www.de-cix.net ------------------------------ DE-CIX Management GmbH Executive Directors: Harald A. Summa and Sebastian Seifert Trade registry: District court (Amtsgericht) Cologne, HRB 51135 Registered office: Lichtstr. 43i, 50825 Cologne
