<div dir="ltr"><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Jan 11, 2013 at 12:18 PM, Jeff Darcy <span dir="ltr"><<a href="mailto:jdarcy@redhat.com" target="_blank">jdarcy@redhat.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div class="im"></div>
My usual answer is: it depends. I've seen cases where using each disk as a separate brick performed better, and I've seen cases where combining them via LVM performed better. There doesn't even seem to be a simple pattern to which will be faster for which workloads, though I'd say brick-per-disk probably wins slightly more often than not. It will also have better failure characteristics than RAID0 - a point also brought up recently by the HDFS folks.<br>
<br>
<a href="http://hortonworks.com/blog/why-not-raid-0-its-about-time-and-snowflakes/" target="_blank">http://hortonworks.com/blog/<u></u>why-not-raid-0-its-about-time-<u></u>and-snowflakes/</a><br>
<br>
They're characteristically wrong about having to wait for the slowest (might be true for their specific workload but not for most others as would be the case for RAID1) but make some other good points.<br>
<br></blockquote><div><br></div><div>Excellent link and I have the same concerns about the larger failure mode and impact of losing a single disk in RAID0/LVM stripe.<br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
You also bring up the issue of bricks needing to be the same size. This is kind of true right now. It won't fail completely, but it also won't distribute files properly and that can lead to premature ENOSPC. However, I expect that to be addressed fairly soon so you might want to consider that as you make your longer-term plans.<div class="">
<div class="h5"><br></div></div></blockquote><div><br></div><div>Even with upcoming support for variable sized bricks, are there merits to carving my current 3TB (best bang for buck) disks into 1TB partitions or PVs (to be concatenated as LV) and use the partitions/LVs as my bricks so I am covered when larger TB disks are available?<br>
<br></div><div>Next question: I know that RHS recommends XFS, and there haven't been any updates to <a href="https://bugzilla.redhat.com/show_bug.cgi?id=838784">https://bugzilla.redhat.com/show_bug.cgi?id=838784</a> since October, but I'd really like to stay with ext4 and perhaps some day convert to btrfs. I will be on 2.6.32-279, which I see is affected, but have there been any recent workarounds?<br>
<br></div></div></div></div>