<div dir="ltr">Thanks Everyone! I'll contact IBM to see what additional information they can provide on this setting. I'll also be sure to test both options to see if there's any noticeable differences in performance. </div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Mar 7, 2017 at 2:23 PM, Cooper F. Nelson <span dir="ltr"><<a href="mailto:cnelson@ucsd.edu" target="_blank">cnelson@ucsd.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">It's a compromise, you are trading interactivity/responsiveness for raw<br>
throughput. It's like the difference between choosing the workstation<br>
or the server timer for the Linux kernel. An android phone and<br>
high-performance SQL server have vastly different use cases.<br>
<br>
I personally use the server (I/O sensitive) configuration whenever<br>
possible on Linux servers. However, as I've learned from Michal and<br>
crew, things aren't always what they seem when comes to multi-threaded<br>
workloads, I/O and caching.<br>
<br>
-Coop<br>
<div class="HOEnZb"><div class="h5"><br>
On 3/7/2017 6:04 AM, Brian Hennigar wrote:<br>
> "Select this choice to determine how to balance between I/O bandwidth<br>
> and balanced workload. Choosing I/O sensitive will get higher I/O<br>
> bandwidth when expansion cards are used. Choosing Balanced will allow<br>
> enough frequency for the workload while the microprocessor cores are idle."<br>
><br>
><br>
> Does anyone know if Suricata would benefit from having the I/O Sensitive<br>
> option enabled?<br>
><br>
><br>
<br>
<br>
</div></div><span class="HOEnZb"><font color="#888888">--<br>
Cooper Nelson<br>
Network Security Analyst<br>
UCSD ITS Security Team<br>
<a href="mailto:cnelson@ucsd.edu">cnelson@ucsd.edu</a> x41042<br>
<br>
</font></span></blockquote></div><br></div>