This archive is retained to ensure existing URLs remain functional. It will not contain any emails sent to this mailing list after July 1, 2024. For all messages, including those sent before and after this date, please visit the new location of the archive at https://mailman.ripe.net/archives/list/[email protected]/
[ncc-services-wg] New on RIPE Labs: A RIPE Atlas Probe for every RIPE NCC Member
- Previous message (by thread): [ncc-services-wg] New on RIPE Labs: A RIPE Atlas Probe for every RIPE NCC Member
- Next message (by thread): [ncc-services-wg] New on RIPE Labs: A RIPE Atlas Probe for every RIPE NCC Member
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Henk Uijterwaal
henk at uijterwaal.nl
Thu Nov 3 08:21:56 CET 2011
On 02/11/2011 15:32, Hank Nussbacher wrote: > At 09:52 02/11/2011 +0100, Mirjam Kuehne wrote: > >> [Apologies for duplicate mails] >> >> Dear colleagues, > > 2.5MEuro for a "nice to have" system? What crucial Internet problem are you > trying to solve with this amount of money? I asked myself a similar question. 20,000 probes translates roughly to 1 per active AS, 0.25 per active prefix or 2 per current member. Let's assume for a second that we indeed install those 20,000 probes equally distributed amongst the membership. Costs are now 2.5M/8000 = 300 euro/member. Now look at my current provider. They will have 2 probes installed in a network with O(100,000) customers. That is by any standard way too low to provide any useful information in case of an operational problem. It is also way too low to do any meaningful random sampling of performance that the average customer will see. In short, one will collect lots of data but it is unclear what operational use it will have. Now turn it around: let's say we want to install the probes such that meaningful information about an AS/prefix/member can be obtained. That means installing O(100) probes per unit. But 20000/100=200, so 7800 members will not benefit from this service. In this case, you'd think that the 200 members using the service pay for it (and the others don't). However, 2.5M/200 = 12,500 euro/participant. Looking at the TTM service, I seriously doubt that there will be 200 members willing to pay for this, plus that providing a service for only a small group is not what the RIPE NCC is supposed to be doing. Either way, I have to agree with Hank that the usefulness of this project is very limited. Henk -- ------------------------------------------------------------------------------ Henk Uijterwaal Email: henk(at)uijterwaal.nl http://www.uijterwaal.nl Phone: +31.6.55861746 ------------------------------------------------------------------------------ There appears to have been a collective retreat from reality that day. (John Glanfield, on an engineering project)
- Previous message (by thread): [ncc-services-wg] New on RIPE Labs: A RIPE Atlas Probe for every RIPE NCC Member
- Next message (by thread): [ncc-services-wg] New on RIPE Labs: A RIPE Atlas Probe for every RIPE NCC Member
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]