<div dir="ltr">Hi Diep,<div><br></div><div>You can check it out in the slide from Red Hat or conduct the benchmark by yourself which simulates the high-load environment. I saw the positive effects of changing the swappiness and the scheduler for block devices, for example. I will share the benchmark results when I have new storage for testing.</div>
<div><br></div><div>FYI, the enterprise-storage profile for "tuned" will change most of the options mentioned by Lanning as well as other options such as elevator.</div><div><br></div><div>Regards,</div><div>Cuong</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Dec 5, 2013 at 3:12 PM, Diep Pham Van <span dir="ltr"><<a href="mailto:imeo@favadi.com" target="_blank">imeo@favadi.com</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 Thu, 5 Dec 2013 14:32:32 +0900<br>
Nguyen Viet Cuong <<a href="mailto:mrcuongnv@gmail.com">mrcuongnv@gmail.com</a>> wrote:<br>
<br>
> If you use RHEL 6 or the like, you should consider to install tuned &<br>
> tuned-utils and activate the enterprise-storage profile as the baseline<br>
> before going to more advanced tuning.<br>
</div>Can you provide some benchmark result before and after active<br>
enterprise-storage profile?<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Diep Pham Van <<a href="mailto:imeo@favadi.com">imeo@favadi.com</a>><br>
</font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br>Nguyen Viet Cuong<br>
</div>