CDN server log

Ryan Malayter malayter at gmail.com
Sat May 18 16:21:49 UTC 2013


Djamel, 

If you are looking for a CDN log trace to do academic research work on say, caching algorithms, please be straightforward about your needs and someone (including myself) might be able to help.

If your purposes are commercial, asking for free data won't likely get you far. If you're trying to turn the data into money expect to pay someone for it.



On May 16, 2013, at 10:33 AM, Michal Krsek <michal at krsek.cz> wrote:

> Hi Djamel,
> I'm not sure what you are looking for.
> 
> There is variety of CDN content and popularity is being driven by users and designers.
> 
> If you have CDN that serves pictures, you get most hits on "design pictures", for paid VoD, you get most hits on free trailers. For CatchTV tup you get most hits on new arrivals of popular content. It also depends on geo distribution. Global CDNs get different coverage than regional ones. For live transmissions, you get a lot of content when covering big sports events.
> 
> For adult based content CDN ... you can imagine ...
> 
> Just talking in general, having no permission to provide any log.
> 
>                        With kind regards
>                                Michal
> 
> 
> Dne 16.5.2013 15:16, Djamel Sadok napsal(a):
>> Hi Pete,
>> 
>> I do not use a CDN I am only interested in analyzing content popularity in
>> logs. These could be anonymized.
>> 
>> Djamel
>> 
>> 
>> 
>> On Wed, May 15, 2013 at 3:55 PM, Pete Mastin <pmastin at internap.com> wrote:
>> 
>>> Hi djamel.  If I understand your question - you should take a look at what
>>> sawmill offers. Many of our clients use this product to analyze our cdn
>>> produced logs.
>>> 
>>> http://www.sawmill.net/
>>> 
>>> 
>>> 
>>> Sent from my iPhone
>>> 
>>> On May 15, 2013, at 10:30 AM, "Djamel Sadok" <jamel at cin.ufpe.br> wrote:
>>> 
>>>> Hi,
>>>> 
>>>> Anyone knows of any public CDN server log trace. I am looking for object
>>>> popularity, hit rate information, ...
>>>> 
>>>> Thanks, Djamel
> 
> 




More information about the NANOG mailing list