Re: [gaia] Heterogeneity in network capacity: growing?

Nick Feamster <feamster@CS.Princeton.EDU> Tue, 01 March 2016 15:10 UTC

Return-Path: <feamster@cs.princeton.edu>
X-Original-To: gaia@ietfa.amsl.com
Delivered-To: gaia@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 049741B2CCE for <gaia@ietfa.amsl.com>; Tue, 1 Mar 2016 07:10:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.207
X-Spam-Level:
X-Spam-Status: No, score=-4.207 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 76-M52JrDzk2 for <gaia@ietfa.amsl.com>; Tue, 1 Mar 2016 07:10:38 -0800 (PST)
Received: from greenlight.cs.princeton.edu (greenlight.CS.Princeton.EDU [128.112.136.72]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A6A581B2CCF for <gaia@irtf.org>; Tue, 1 Mar 2016 07:10:38 -0800 (PST)
Received: from nicks-mbp.lan (pool-173-71-96-150.cmdnnj.fios.verizon.net [173.71.96.150]) (authenticated bits=0) by greenlight.cs.princeton.edu (8.14.4/8.14.4) with ESMTP id u21FARSd004745 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Tue, 1 Mar 2016 10:10:30 -0500
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2104\))
From: Nick Feamster <feamster@CS.Princeton.EDU>
In-Reply-To: <56D5A485.6080503@ripe.net>
Date: Tue, 1 Mar 2016 10:10:26 -0500
Content-Transfer-Encoding: quoted-printable
Message-Id: <CC7388DC-F460-4F9B-BD5F-5E913E95223B@cs.princeton.edu>
References: <32116_1456700693_56D37D15_32116_453_1_CF23F499-41D1-4DA6-A306-8924DC4F4DB4@ifi.uio.no> <41954A84-EFF0-4107-9859-E60AAC65285C@aalto.fi> <56F64461-28A0-4E5E-9923-D177B663EFB7@ifi.uio.no> <e3a59b97b74543a7960911bc155e8830@rew09926dag05b.domain1.systemhost.net> <E41521ED-AA4F-4AE5-BA02-6BD7B7DFC621@ifi.uio.no> <EE9E048C-DD05-4DB3-929F-DF337355DC90@isoc.org> <56D59D94.2090109@ripe.net> <44250FD7-84FD-4A9F-B133-8B6682BF278F@ifi.uio.no> <56D5A485.6080503@ripe.net>
To: Mirjam Kuehne <mir@ripe.net>
X-Mailer: Apple Mail (2.2104)
Archived-At: <http://mailarchive.ietf.org/arch/msg/gaia/bAEhkIbRIzmF880BU-zXDk1rWsw>
Cc: gaia@irtf.org, Michael Welzl <michawe@ifi.uio.no>
Subject: Re: [gaia] Heterogeneity in network capacity: growing?
X-BeenThere: gaia@irtf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: Global Access to the Internet for All <gaia.irtf.org>
List-Unsubscribe: <https://www.irtf.org/mailman/options/gaia>, <mailto:gaia-request@irtf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/gaia/>
List-Post: <mailto:gaia@irtf.org>
List-Help: <mailto:gaia-request@irtf.org?subject=help>
List-Subscribe: <https://www.irtf.org/mailman/listinfo/gaia>, <mailto:gaia-request@irtf.org?subject=subscribe>
X-List-Received-Date: Tue, 01 Mar 2016 15:10:42 -0000

Folks may also be interested in the BISmark project:
http://projectbismark.net/

We measure throughput from hundreds of home networks around the world in about 30 countries.  All of that data is available at http://uploads.projectbismark.net/

We also have a mobile throughput application:
http://play.google.com/store/apps/details?id=com.num

We’ve used these to study throughput in the US, South Africa, and Pakistan. In South Africa and Pakistan, we looked at both fixed line and mobile throughput.  Here are the relevant papers:

* US: http://goo.gl/p48tDr
* South Africa: http://goo.gl/kONyHu
* Pakistan: http://goo.gl/9RdzXK

-Nick

> On Mar 1, 2016, at 9:17 AM, Mirjam Kuehne <mir@ripe.net> wrote:
> 
> Hi Michael,
> 
> You are right: RIPE Atlas doesn't measure bandwidth but I thought
> measuring latency could add another aspect to the question.
> 
> Kind regards,
> Mirjam
> 
> 
> On 1/3/16 15:04, Michael Welzl wrote:
>> Thanks a lot for that pointer too!
>> I know about the RIPE atlas and do find it very interesting.
>> 
>> However, probably I'm missing something - how is the traceroute dataset relevant for what I'm asking?
>> My question is about link bandwidth...
>> 
>> 
>>> On 01 Mar 2016, at 14:48, Mirjam Kuehne <mir@ripe.net> wrote:
>>> 
>>> Hi,
>>> 
>>> Thanks to Jane Coffin for pointing me to this discussion.
>>> 
>>> You might also want to have a look at the RIPE Atlas system, a worldwide
>>> active measurements network: https://atlas.ripe.net/
>>> 
>>> While we don't actually measure bandwidth, the thousands of RIPE Atlas
>>> measurement probes regularly send pings and traceroutes (and also SSL,
>>> DNS queries) to certain pre-defined destinations on the Internet.
>>> Especially the traceroute results could be interesting for the questions
>>> below.
>>> 
>>> The measurement results are public and are used by many researchers and
>>> operators for various use cases. Have a look at this blog for some
>>> inspiration: https://labs.ripe.net/atlas
>>> 
>>> If you have any questions, please don't hesitate to contact me directly.
>>> 
>>> Kind regards,
>>> Mirjam Kuehne
>>> RIPE NCC
>>> 
>>>> 
>>>> On 2/29/16, 2:59 PM, "gaia on behalf of Michael Welzl" <gaia-bounces@irtf.org on behalf of michawe@ifi.uio.no> wrote:
>>>> 
>>>>> Great, thanks!
>>>>> 
>>>>> Good thinking about Akamai’s report - of course! I should have looked at that first
>>>>> 
>>>>> 
>>>>>> On 29. feb. 2016, at 19.22, achilles.petras@bt.com wrote:
>>>>>> 
>>>>>> Hi Michael
>>>>>> 
>>>>>> A good starting point is Ofcom's infrastructure annual report; here is the 2015 one:
>>>>>> http://stakeholders.ofcom.org.uk/binaries/research/infrastructure/2015/downloads/connected_nations2015.pdf
>>>>>> 
>>>>>> You will find several distribution graphs like the one in Figure 4 that compares the monthly usage to the access speed range for 2014 and 2015.
>>>>>> 
>>>>>> You could also find the reports from previous years to see how things have changed over time:
>>>>>> http://stakeholders.ofcom.org.uk/market-data-research/market-data/infrastructure/?a=0
>>>>>> 
>>>>>> For a more global point of view, you could check Akamai's State of the Internet Report:
>>>>>> https://www.stateoftheinternet.com/resources-connectivity-2015-q3-state-of-the-internet-report.html
>>>>>> 
>>>>>> or play with their visualisations:
>>>>>> https://www.stateoftheinternet.com/trends-visualizations-connectivity-global-heat-map-internet-speeds-broadband-adoption.html
>>>>>> 
>>>>>> Hope this helps
>>>>>> Achilles
>>>>>> 
>>>>>> 
>>>>>> -----Original Message-----
>>>>>> From: gaia [mailto:gaia-bounces@irtf.org] On Behalf Of Michael Welzl
>>>>>> Sent: 29 February 2016 08:16
>>>>>> To: Manner Jukka <jukka.manner@aalto.fi>
>>>>>> Cc: gaia@irtf.org
>>>>>> Subject: Re: [gaia] Heterogeneity in network capacity: growing?
>>>>>> 
>>>>>> First,
>>>>>> 
>>>>>> Apologies for asking first formulating a question in two ways with contradictory outcomes (yes/no) and then saying "I guess the answer is no" - that was confusing!  :)   I was tired when I wrote it...
>>>>>> so I meant the second question, i.e. my guess was also that heterogeneity *is* growing.
>>>>>> 
>>>>>> How would you calculate if there was data: I guess what I'm interested in is bandwidth statistics: how often do we see a bottleneck of X bit/s? The Internet-wide distribution of X is what I'm looking for...
>>>>>> 
>>>>>> Depending on area/market: I'd also be happy to have X for India, X for Europe, X for ... whatever. It's a start!
>>>>>> Ideally citable - some publicly available study I could refer to.
>>>>>> 
>>>>>> Cheers,
>>>>>> Michael
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>>> On 29 Feb 2016, at 08:33, Manner Jukka <jukka.manner@aalto.fi> wrote:
>>>>>>> 
>>>>>>> Hi,
>>>>>>> 
>>>>>>> I would say yes, they are globally, but it also depends on what area/market you are talking about. How would you calculate it if there was data? I have data, but not sure how to calculate this.
>>>>>>> 
>>>>>>> cheers,
>>>>>>> Jukka
>>>>>>> 
>>>>>>>> On 29 Feb 2016, at 01:04, Michael Welzl <michawe@ifi.uio.no> wrote:
>>>>>>>> 
>>>>>>>> Hi,
>>>>>>>> 
>>>>>>>> I guess this is probably the right group to ask: is the heterogeneity of link bandwidths growing?
>>>>>>>> Another way of asking this is: across the whole Internet, do old links roughly get removed as quickly as capacities grow?
>>>>>>>> 
>>>>>>>> I guess the answer is no - but are there any good citable references for this? Data?
>>>>>>>> 
>>>>>>>> Cheers,
>>>>>>>> Michael
>>>>>>>> 
>>>>>>>> _______________________________________________
>>>>>>>> gaia mailing list
>>>>>>>> gaia@irtf.org
>>>>>>>> https://www.irtf.org/mailman/listinfo/gaia
>>>>>>> 
>>>>>> 
>>>>>> _______________________________________________
>>>>>> gaia mailing list
>>>>>> gaia@irtf.org
>>>>>> https://www.irtf.org/mailman/listinfo/gaia
>>>>> 
>>>>> _______________________________________________
>>>>> gaia mailing list
>>>>> gaia@irtf.org
>>>>> https://www.irtf.org/mailman/listinfo/gaia
>>> 
>>> _______________________________________________
>>> gaia mailing list
>>> gaia@irtf.org
>>> https://www.irtf.org/mailman/listinfo/gaia
>> 
> 
> _______________________________________________
> gaia mailing list
> gaia@irtf.org
> https://www.irtf.org/mailman/listinfo/gaia