At 15:31 03/11/2011 +0100, Daniel Karrenberg wrote:
Henk,
good to hear from you again.
RIPE Atlas is not intended for intra-AS measurements. Operators generally have their networks very well instrumented. RIPE Atlas is about situational awareness about the inter-AS mesh. Of course we want to have a number of probes inside big ASes, but most stub ASes are nicely covered with one or two. We would not provide other than incidental intra-AS measuremenets from the proposed funding unless the operator concerned provides additional funding and the results will be both available and useful for the community.
So I am indeed proposing to distribute the probes and benefits as equally among the membership as possible. And the cost of 300 EUR for each member that you quote is for the deployment and two years of operations, so it comes down to EUR150/member/year for the first two years.
I believe only 1% of the RIPE membership would make weekly use of such probes - you might believe I am way off and feel 10% of the RIPE membership would make use of such probes. Well, there is a simple way to determine whether the membership wants to spend a few million Euro on this - ask them via a poll. If a majority of those responding feel it is important - then you have your mandate. If a majority think otherwise, you don't have a mandate to spend all that money. It is called democracy. -Hank
About the usefulness of RIPE Atlas let me say just this much: At least one active probe in the network of each member would make for an unprecedented level of comprehensive active measurmenets. Products from these measurements are useful for detecting anomalies, monitoring trends and for analysing faults.
Daniel