RE: Bandwidth Growth

[Sameer Khosla]

-Original Message-
From: Curran, David [mailto:david.cur...@windstream.com] 

> The LINX, TORIX, and SIX graphs provided earlier, for example, seem
> relatively flat until the Nov. timeframe at which point they seem to
seek
> a new higher "normal".  Could just be my lack of sleep and my biased
> opinion though.  As one might infer, I'm trying to find evidence
> corroborating my own experiences.  My theory being that all of those
cool
> Internet connected toys (Blue Ray, TVs, Ipads, etc) that hit the
market
> this past year have lead to a substantial increase in residential
Internet
> traffic.  If that were true, I guess I'd expect that enterprise
oriented
> service providers would not see the same uptick as residential
providers.
> 
> Thanks again to all have responded (and anyone else that might still).
>
> David


David, you may also want to try to correlate the addition of
participants to the peering fabric to bandwidth growth.  I know TorIX
shows on their main page the date when new participants joined the
exchange.

http://www.torix.net/

For example, Microsoft jointed on 9/3/10.  You may be able to see a rise
in usage shortly after.

Thanks
Sameer

Re: Bandwidth Growth

[Joe Loiacono]

"Curran, David"  wrote on 04/21/2011 08:52:29 
AM:

> ... it also strikes me that on the aggregate
> the graphs do indeed show a significant increase around the "holidays"
> (the US ones anyway).

Another bias? :-)

Seems Internet participants took some time off for Christmas:

https://stats.linx.net/aggregate.html
http://www.seattleix.net/agg.htm

Joe

Re: Bandwidth Growth

[Curran, David]

 > Date: Thu, 21 Apr 2011 10:36:54 +0800
 > From: Adrian Chadd 
 > Subject: Re: Bandwidth growth
 > To: "Patrick W. Gilmore" 
 > Cc: NANOG list 
 > Message-ID: <20110421023654.ge13...@skywalker.creative.net.au>
 > Content-Type: text/plain; charset=us-ascii
 >
 > If it's a true research project, wouldn't you really be interested in
both
 > evidence for/against? :-)
 >
 > Just my 2c here,
 >
 >
 > Adrian


Touche.  Fortunately not being in the academic world allows me to be
completely biased! :)  But your point is well taken, and I'll take
evidence for either side.  I appreciate all of the responses I've received
thus far.  Especially the links to the IX graphs.  While I agree you can't
account for peering/de-peering, it also strikes me that on the aggregate
the graphs do indeed show a significant increase around the "holidays"
(the US ones anyway).

The LINX, TORIX, and SIX graphs provided earlier, for example, seem
relatively flat until the Nov. timeframe at which point they seem to seek
a new higher "normal".  Could just be my lack of sleep and my biased
opinion though.  As one might infer, I'm trying to find evidence
corroborating my own experiences.  My theory being that all of those cool
Internet connected toys (Blue Ray, TVs, Ipads, etc) that hit the market
this past year have lead to a substantial increase in residential Internet
traffic.  If that were true, I guess I'd expect that enterprise oriented
service providers would not see the same uptick as residential providers.

Thanks again to all have responded (and anyone else that might still).

David

***
The information contained in this message, including attachments, may contain
privileged or confidential information that is intended to be delivered only to 
the
person identified above. If you are not the intended recipient, or the person
responsible for delivering this message to the intended recipient, Windstream 
requests
that you immediately notify the sender and asks that you do not read the 
message or its
attachments, and that you delete them without copying or sending them to anyone 
else.

Re: Bandwidth growth

[Martin Millnert]

On Wed, Apr 20, 2011 at 9:55 PM, Patrick W. Gilmore  wrote:
> On Apr 20, 2011, at 9:35 PM, Curran, David wrote:
>
>> I'm interested in any evidence (even anecdotal) that general Internet usage 
>> (and more importantly, link utilization) has increased at higher rates in 
>> the last 6-12 months than in previous periods.  Any graphs or otherwise 
>> would be greatly appreciated.  The purpose is for an internal research 
>> project and this data will only be used internally and will not be shared, 
>> nor will the sources.
>
> 
> 
> 
> 
> 

Growth unsurprisingly also varies by region:
http://www.msk-ix.ru/eng/traffic.html
It has seen plenty of growth recently.

If any MSK-IX staff reads this, a 3-, 5- or all-year graph would be an
interesting add!

> I don't know if that proves your theory.  And one could argue public IX stats 
> are actually not representative of growth, since many networks move peers to 
> private connections as they grow.  But it is data, and it is available.

Aggregate IX statistics also fail to identify what part of the growth
is due to people moving traffic onto IX:es, from private connections
(transits).  It is certainly data, aggregate data. I wouldn't hang my
heart-lung machine off of it's accuracy in predicting individual
networks short-term traffic developments though, so to speak. :)

Regards,
Martin

Re: Bandwidth growth

[Adrian Chadd]

If it's a true research project, wouldn't you really be interested in both
evidence for/against? :-)

Just my 2c here,


Adrian

On Wed, Apr 20, 2011, Patrick W. Gilmore wrote:
> On Apr 20, 2011, at 9:35 PM, Curran, David wrote:
> 
> > I'm interested in any evidence (even anecdotal) that general Internet usage 
> > (and more importantly, link utilization) has increased at higher rates in 
> > the last 6-12 months than in previous periods.  Any graphs or otherwise 
> > would be greatly appreciated.  The purpose is for an internal research 
> > project and this data will only be used internally and will not be shared, 
> > nor will the sources.
> 
> 
> 
> 
> 
> 
> 
> Etc.
> 
> I don't know if that proves your theory.  And one could argue public IX stats 
> are actually not representative of growth, since many networks move peers to 
> private connections as they grow.  But it is data, and it is available.
> 
> -- 
> TTFN,
> patrick
> 

-- 
- Xenion - http://www.xenion.com.au/ - VPS Hosting - Commercial Squid Support -
- $24/pm+GST entry-level VPSes w/ capped bandwidth charges available in WA -

Re: Bandwidth growth

[Patrick W. Gilmore]

On Apr 20, 2011, at 9:35 PM, Curran, David wrote:

> I'm interested in any evidence (even anecdotal) that general Internet usage 
> (and more importantly, link utilization) has increased at higher rates in the 
> last 6-12 months than in previous periods.  Any graphs or otherwise would be 
> greatly appreciated.  The purpose is for an internal research project and 
> this data will only be used internally and will not be shared, nor will the 
> sources.







Etc.

I don't know if that proves your theory.  And one could argue public IX stats 
are actually not representative of growth, since many networks move peers to 
private connections as they grow.  But it is data, and it is available.

-- 
TTFN,
patrick

Bandwidth growth

[Curran, David]

I'm interested in any evidence (even anecdotal) that general Internet usage 
(and more importantly, link utilization) has increased at higher rates in the 
last 6-12 months than in previous periods.  Any graphs or otherwise would be 
greatly appreciated.  The purpose is for an internal research project and this 
data will only be used internally and will not be shared, nor will the sources.

Thanks in advance,

David Curran  I  New Technology Planning  I  Windstream
O-864.331.7132  I  C-864.905.0522 I 
david.cur...@windstream.com
***
The information contained in this message, including attachments, may contain
privileged or confidential information that is intended to be delivered only to 
the
person identified above. If you are not the intended recipient, or the person
responsible for delivering this message to the intended recipient, Windstream 
requests
that you immediately notify the sender and asks that you do not read the 
message or its
attachments, and that you delete them without copying or sending them to anyone 
else.