<div dir="ltr">On Wed, Sep 10, 2008 at 3:16 PM, Ed Loach <span dir="ltr"><<a href="mailto:ed@loach.me.uk">ed@loach.me.uk</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><br></div><br>
I think this is an improvement; it means we seem to have clients<br>
ready to start servicing the level 1 and 2 requests almost as soon<br>
as they come in (when all is running smoothly that is), and the<br>
level 4 requests are getting dealt with without the upload queue<br>
filling and causing everything to choke up.<br>
<font color="#888888"></font></blockquote><div><br>But this means my CPUs are sitting idle with no work to do, and I'm paying for their power and heat to do nothing. I'd rather see them doing preemptive renders of highly-edited areas or rendering those prio4 requests than sitting idle.<br>
<br>How many requests per second is the TAH server able to handle? It doesn't seem like shuffling this sort of data around would be too much work, but I didn't pay attention to those huge discussions about tileset-per-file earlier. Did we cover this issue then?<br>
</div></div><br></div>