Re: [gnso-ff-pdp-may08] fast flux numbers lately

[Martin Hall <martinh@xxxxxxxxxxxxxxx>]

I've added our numbers for the past 3 months to Greg's spreadsheet.
I'm working separately on charting this and other data for a draft to  
insert in the report.
We have rather a lot of data that's taking longer than expected to run  
our complete analytics on.
Looking at the ff domains/tld with the per 10,000 perspective, we see  
the following TLDs with 1 or more per 10,000:
cn 19.55 (24,171)

Attachment: FF stats by TLD - Arbor and Karmasphere.xls
Description: MS-Excel spreadsheet

com 2.62 (20,488)
su 6.1 (42)
bz 4.37 (19)
ec 1.23 (2)
net 1.36 (1617)
gs 19.05 (16)
asia 1.0 (21)

Overall this concurs with Arbor data and Joe's observation about .com  
& .cn. One observation to add: in the past 60 days, we've seen a  
significant uptick in .cn domains fluxing in both real and factored  
terms.
Martin

On Dec 2, 2008, at 2:43 PM, Joe St Sauver wrote:

> So computing D2/C2*10000 (ff domains observed by Arbor/domains in the
> TLD*10000) as shown on Greg's spreadsheet and then formatting for  
> three
> decimal places, I'm seeing:
>
> com  2.151 (n=16818)
> cn   5.170 (n= 6393)
> ru   1.010 (n=  155)
> su   7.548 (n=   52)
> bz   3.218 (n=   14)
> gs  14.286 (n=   12)
> ec   3.077 (n=    5)
>
> Everything else is <1.0. At least from my point of view, dot com and
> dot cn continue to strike me as the primary FF problem locii.
>
> It will be interesting to see if the Karmasphere numbers are radically
> different.
>
> Regards,
>
> Joe
--
Martin Hall
skype: martin-hall
+1-408-838-2890