Re: [Asrg] hashcash, was My take on e-postage

Jonathan Morton wrote:


> > Here's a table:
> >
> > Age (days)   IP Count/cum%  Hit Count/cum%
> >          0      67974/ 5.0     174862/25.4
> >          1      65816/ 9.8     184589/52.1
> >          2      54233/13.8     118338/69.3
> >          3      48207/17.4      50991/76.7
> >          4      38517/20.2      45567/83.3
> I don't get it.  Why is the hit count for some days so much lower than 
> the host count?
The detector is running on a multitude of dmoains with vastly higher 
flows in total, whereas the hit count above is on our own production 
mail service.
> Anyway, assuming your figures are correct, that indicates that 2/3rds of 
> the hosts are rolling around to the beginning of the list again, because 
> they're being detected as new immediately after being expired.  That in 
> turn seems to suggest 25k new zombies per day, rather than 70k.
Not quite - there's some information you're missing.  Minimum 
redetection interval is 48 hours, yet this is on last detection.  On the 
other hand, the numbers for previous days isn't going down by much. 
That first number (70K) is continuously rising.  A month or two ago it 
was around 15-20K, a few months before that, it was 5kish.  The 
redetection interval hasn't changed.
> The 
> total zombie pool is probably a more reliable figure, which I'll round 
> up to 1500k.
Dynamic IP detections (this thing makes no distinction) would tend to 
push it back down.
When the expiration interval was four months, the total was around 
2200K. The reduction of 4mos -> 1mos resulted in an immediate 
2200K->1000k reduction, but only lost 3% of the detections.

_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg