Re: [ris-int] Tools using the RIS
Arife and I just sat through a bof on BGP analysis where a few tools using RIS and/or RV data were presented. I've no idea how useful those tools actually are but if they are only presented at NANOG, I doubt that many people in the RIPE region will hear that tehy are available, let alone test drive them.
Random idea: shouldn't we make a page on the ris website with links to all the tools that we know of.
That's one thing, but what I think the RIPE community (operators) would really like is for the RIS service to be expanded with tools we consider best, as was done with BGPlay. Operators don't have time to investigate the work done by researchers in various universities, they need hands on tools working on the full set of RIS data. With the large number of peers RIS has now, any query via the RIS cgis yields an overwhelming amount of data. RIS could sure do with some intermediate, higher level, overview of the results. BGPinspect e.g. starts with a nice summary (graphical and in text) before listing the raw results. (http://bgpinspect.merit.edu/) -- Rene
At 04:56 25/10/2005, Rene Wilhelm wrote:
Arife and I just sat through a bof on BGP analysis where a few tools using RIS and/or RV data were presented. I've no idea how useful those tools actually are but if they are only presented at NANOG, I doubt that many people in the RIPE region will hear that tehy are available, let alone test drive them.
Random idea: shouldn't we make a page on the ris website with links to all the tools that we know of.
That's one thing, but what I think the RIPE community (operators) would really like is for the RIS service to be expanded with tools we consider best, as was done with BGPlay. Operators don't have time to investigate the work done by researchers in various universities, they need hands on tools working on the full set of RIS data.
In principle I agree, but I do see two problems here: * We're not operators, so it will be hard for us to judge if a tool is valuable or not. And I certainly don't want to label tools as "not useful". * If we support a tool, then we should be fairly certain that it will exist for a while. I.e. it is not a student's project where the student has gone on, that suddenly disappears or is not supported. For BGPlay we did check that the Rome group would continue to develop it. So, I think we should be neutral here and just list what is available. However, when we insert a link, we can always ask the authors to post to the lists, introduce the tool and ask for feedback. If there is a good product, people will pick it up. Henk
With the large number of peers RIS has now, any query via the RIS cgis yields an overwhelming amount of data. RIS could sure do with some intermediate, higher level, overview of the results.
BGPinspect e.g. starts with a nice summary (graphical and in text) before listing the raw results. (http://bgpinspect.merit.edu/)
-- Rene
------------------------------------------------------------------------------ Henk Uijterwaal Email: henk.uijterwaal(at)ripe.net RIPE Network Coordination Centre http://www.amsterdamned.org/~henk P.O.Box 10096 Singel 258 Phone: +31.20.5354414 1001 EB Amsterdam 1016 AB Amsterdam Fax: +31.20.5354445 The Netherlands The Netherlands Mobile: +31.6.55861746 ------------------------------------------------------------------------------ Look here junior, don't you be so happy. And for Heaven's sake, don't you be so sad. (Tom Verlaine)
On Tue, 25 Oct 2005, Henk Uijterwaal wrote:
That's one thing, but what I think the RIPE community (operators) would really like is for the RIS service to be expanded with tools we consider best, as was done with BGPlay. Operators don't have time to investigate the work done by researchers in various universities, they need hands on tools working on the full set of RIS data.
In principle I agree, but I do see two problems here:
* We're not operators, so it will be hard for us to judge if a tool is valuable or not. And I certainly don't want to label tools as "not useful".
* If we support a tool, then we should be fairly certain that it will exist for a while. I.e. it is not a student's project where the student has gone on, that suddenly disappears or is not supported. For BGPlay we did check that the Rome group would continue to develop it.
So, I think we should be neutral here and just list what is available.
Of course, listing the tools on the RIS website is good, as it might help raise awareness. And not every conceivable tool would be in scope for RIS to support directly. But with a "user" hat on, I'd say RIS could use something which gives summarized/aggregate information. We have so many peers on more than a dozen RRCs that users easily can get lost. Unless you are looking for something specific, it can be hard to mine all the available data. I see your point regarding continued support, so maybe RIPE NCC should take inspiration from the work done in the research community (e.g. BGPinspect) and use that in expanding the available tools for looking at RIS data. Another idea might be to run some sort of survey of our own, how people use RIS, what they'd like to see added, worked on. Post it to routing-wg@ripe.net to reach all in the RIPE community (most of nanog just seem to default to routeviews) -- Rene
participants (2)
-
Henk Uijterwaal -
Rene Wilhelm