<div dir="ltr">Thanks for sharing this information guys! We'll test the PipeBackend capabilities in the coming days.<div><br></div><div style><font color="#999999">Ron</font></div></div><div class="gmail_extra"><br><br>
<div class="gmail_quote">On Tue, Mar 19, 2013 at 11:56 PM, bert hubert <span dir="ltr"><<a href="mailto:bert.hubert@netherlabs.nl" target="_blank">bert.hubert@netherlabs.nl</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><div><div class="im"><div>On Mar 19, 2013, at 10:41 PM, Ron Tsoref wrote:</div><br><blockquote type="cite"><div dir="ltr">The PipeBackend seems easy to implement. Does anyone actually use a PipeBackend in production and can share some general performance information? Is it much slower than other backends?<div class="gmail_extra">
<br></div></div></blockquote><div><br></div></div><div>We've been able to squeeze 50000 qps out of a pipe backend. On a philosophical note, pipes are likely to be faster than TCP/IP, and SQL marshalling/unmarshaling is not free either.</div>
<div><br></div><div>People associate 'text based' with slow, but most SQL protocols are just as parsed, or even more so.</div><div><br></div><div>The pipe backend does have a performance bottleneck in 3.2 if you specify a timeout, see <a href="http://wiki.powerdns.com/trac/ticket/661" target="_blank">http://wiki.powerdns.com/trac/ticket/661</a></div>
<span class="HOEnZb"><font color="#888888"><div><br></div><div><span style="white-space:pre-wrap"> </span>Bert</div><div><br></div></font></span></div></div></blockquote></div><br></div>