On Mon, Apr 26, 2010 at 5:52 PM, Platonides <span dir="ltr">&lt;<a href="mailto:Platonides@gmail.com">Platonides@gmail.com</a>&gt;</span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="im">Anthony wrote:<br>
&gt; What kind of space needs are we talking about?<br>
<br>
</div>100k requests per second.<br>
Assuming that an url is 50 bytes on average, that&#39;s 432 GB per day (the<br>
usual apache log line is about 1.5 times that).<br></blockquote><div><br>Seems reasonable.  For 3 days of access that&#39;s 18 gigs per server over 70 servers.<br><br>And that&#39;s without compression, and 50 bytes seems awfully long for a URL.<br>
 </div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="im">&gt;     What if your referer was your facebook personal page leaking your full<br>
&gt;     real name?<br>
&gt;<br>
&gt; And what if you&#39;re in the sample?  I find it quite inappropriate that<br>
&gt; even sampled data like this is being released.<br>
<br>
</div>The referer is not stored anywhere.<br></blockquote><div><br>Well, that&#39;s good to hear.  What exactly is contained in the sampled data which is being released?  We&#39;ve heard what&#39;s in the 1/10th sample Mr. Priedhorsky is getting, but what about the rest?<br>
</div></div>