<br><br><div class="gmail_quote">On Mon, Oct 29, 2012 at 1:22 PM, Julien Danjou <span dir="ltr"><<a href="mailto:julien@danjou.info" target="_blank">julien@danjou.info</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Mon, Oct 29 2012, Doug Hellmann wrote:<br>
<br>
>> So the idea is to mitigate this by sharing polling cycles where possible.<br>
>><br>
>> Say the metering publisher needs to sample every 10 minutes, whereas a<br>
>> metrics publisher needs a more fine-grain, say every 2 minutes. In that<br>
>> case, the yield from every fifth polling cycle will be shared and published<br>
>> by both.<br>
>><br>
><br>
> Sure, that makes sense. What do we do if the user configures a pollster to<br>
> run every 10 seconds?<br>
<br>
</div>The configuration semantic has to change in this regard. Users shouldn't<br>
configure pollster interval time, but publishers frequencies.<br></blockquote><div><br></div><div>That doesn't really solve the problem, though, because they can still say they want the data published every 10 seconds and then we have to compute it that often.</div>
<div><br></div><div>Doug</div><div> </div></div>