Hi all, two quick questions: Is there any way to retrieve BGP data (e.g. table dumps, updates, ...) such that i.) the data is not already available in the RIPE RIS, Routeviews, PCH, Isolario, or BGPmon projects and ii.) it is not necessary to query a Looking Glass to death (e.g. get all neighbors, for each neighbor get all prefixes, for each prefix get all announcements)? Are there any traceroute (or related) projects that currently still make their data publically available except for MLAB, RIPE Atlas, Caida, the Bismark Projekt, the portolan project? In addition, if you know somebody that knows somebody (...) that probably might be capable of giving me access to an IXP Route Server dump, a set of traceroutes that is not published publically yet, or something related I would *really* appreciate if you could forward me the contact! :) All data is supposed to be used in a research context, more specific my master thesis, therefore I would also accept annonymized data as long as it still contains useful topology information. Best regards, Lars
Is this some sort of BGP AS Path Visualization like what ThousandEyes are doing? Sent from ProtonMail mobile -------- Original Message -------- On Apr 26, 2018, 4:46 PM, Lars Prehn wrote:
Hi all,
two quick questions:
Is there any way to retrieve BGP data (e.g. table dumps, updates, ...) such that i.) the data is not already available in the RIPE RIS, Routeviews, PCH, Isolario, or BGPmon projects and ii.) it is not necessary to query a Looking Glass to death (e.g. get all neighbors, for each neighbor get all prefixes, for each prefix get all announcements)?
Are there any traceroute (or related) projects that currently still make their data publically available except for MLAB, RIPE Atlas, Caida, the Bismark Projekt, the portolan project?
In addition, if you know somebody that knows somebody (...) that probably might be capable of giving me access to an IXP Route Server dump, a set of traceroutes that is not published publically yet, or something related I would *really* appreciate if you could forward me the contact! :) All data is supposed to be used in a research context, more specific my master thesis, therefore I would also accept annonymized data as long as it still contains useful topology information.
Best regards,
Lars
No. Visualizing AS paths, as such, is not planned within the scope of my thesis :) However, I'm really interested in getting an accurate snapshot of the current Internet's AS-level topology. The topology-related data ThousandEyes collects would fit my needs perfectly. If anyone may be able to provide similar data for academic research purposes I would really appreciate receiving a mail. Best regards, Lars Am 26.04.18 um 20:33 schrieb Timothy Manito:
Is this some sort of BGP AS Path Visualization like what ThousandEyes are doing?
Sent from ProtonMail mobile
-------- Original Message -------- On Apr 26, 2018, 4:46 PM, Lars Prehn < lprehn@inet.tu-berlin.de> wrote:
Hi all,
two quick questions:
Is there any way to retrieve BGP data (e.g. table dumps, updates, ...) such that i.) the data is not already available in the RIPE RIS, Routeviews, PCH, Isolario, or BGPmon projects and ii.) it is not necessary to query a Looking Glass to death (e.g. get all neighbors, for each neighbor get all prefixes, for each prefix get all announcements)?
Are there any traceroute (or related) projects that currently still make their data publically available except for MLAB, RIPE Atlas, Caida, the Bismark Projekt, the portolan project?
In addition, if you know somebody that knows somebody (...) that probably might be capable of giving me access to an IXP Route Server dump, a set of traceroutes that is not published publically yet, or something related I would *really* appreciate if you could forward me the contact! :) All data is supposed to be used in a research context, more specific my master thesis, therefore I would also accept annonymized data as long as it still contains useful topology information.
Best regards,
Lars
On Thu, 26 Apr 2018, Lars Prehn wrote:
However, I'm really interested in getting an accurate snapshot of the current Internet's AS-level topology. The topology-related data ThousandEyes collects would fit my needs perfectly. If anyone may be able to provide similar data for academic research purposes I would really appreciate receiving a mail.
Hello Lars Checking to see what this research activity produced could be helpful. Towards an Accurate, Geo-Aware, PoP-Level Perspective of the Internet's Inter-AS Connectivity https://nsf.gov/awardsearch/showAward?AWD_ID=1320977 Steve
Geo-positioning, in general sucks. Internet topology has very little to do with geo-data, however, i would highly recommend this guys https://www.ipip.net/ On Fri, Apr 27, 2018 at 6:10 PM, Steven G. Huter <sghuter@nsrc.org> wrote:
On Thu, 26 Apr 2018, Lars Prehn wrote:
However, I'm really interested in getting an accurate snapshot of the
current Internet's AS-level topology. The topology-related data ThousandEyes collects would fit my needs perfectly. If anyone may be able to provide similar data for academic research purposes I would really appreciate receiving a mail.
Hello Lars
Checking to see what this research activity produced could be helpful.
Towards an Accurate, Geo-Aware, PoP-Level Perspective of the Internet's Inter-AS Connectivity
https://nsf.gov/awardsearch/showAward?AWD_ID=1320977
Steve
-- Alexander Lyamin, VP & Founder Qrator <http://qrator.net/>* Labs CZ * office: +420 602 558 144 <++420+602+558+144> mob: +420 774 303 807 <++420+774+303+807> skype: melanor9 mailto: la@qrator.net
On 27/04/18 04:33, Timothy Manito via NANOG wrote:
Is this some sort of BGP AS Path Visualization like what ThousandEyes are doing? I wrote something like that last year using all AS15169 peerings, sourcing data from BMP, then rendering out all the various paths just using graphviz.
The most useful thing was the version I made which did that for direct peers and their single-homed (at least as far as we could tell) customers. So many networks have a couple of prefixes they forgot to advertise on peerings, a transit they forgot to v6 enable, or an "I'm sure the bosses didn't know about this one" v6 transit path. The most amusing graph was probably Akamai, since they reuse AS20940 all over the planet for distinct nodes. That one probably qualifies as modern art.
participants (5)
-
Alexander Lyamin -
Julien Goodwin -
Lars Prehn -
Steven G. Huter -
Timothy Manito