<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
On 02/17/2017 10:13 AM, Gambit15 wrote:<br>
<blockquote
cite="mid:CAEfk3RXJjNz8EQCCYPfwdtwHmtUoMOBSp2KhqgbmE5wo851HSA@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
RAID is not an option, JBOD with EC will be used.<br>
</blockquote>
<div><br>
</div>
<div>Any particular reason for this, other than maximising
space by avoiding two layers of RAID/redundancy?<br>
</div>
<div>Local RAID would be far simpler & quicker for
replacing failed drives, and it would greatly reduce the
number of bricks & load on Gluster.<br>
<br>
</div>
<div>We use RAID volumes for our bricks, and the benefits of
simplified management far outweigh the costs of a little
lost capacity.<br>
</div>
<div><br>
</div>
<div>D<br>
</div>
</div>
</div>
</div>
</blockquote>
<br>
This is as much of a question as a comment.<br>
<br>
My impression is that distributed filesystems like Gluster shine
where the number if bricks is close to the number of servers and
both of those numbers are as large as possible.<br>
So the ideal solution would be 90 disks as 90 bricks on 90 servers.<br>
<br>
This would be hard to do in practice but the point of Gluster is to
try and spread the load and potential failures over a large surface.<br>
<br>
Putting all the disks into a big RAID array and then just
duplicating that for redundancy is not much better than using
something like DRBD which would likely perform faster but be less
scaleable.<br>
In the end with big RAID arrays and fewer servers you have a smaller
surface to absorb failures.<br>
<br>
Over the years I have seen raid systems fail because users put them
in and forget about them and then see system failures becasue they
did not monitor the raid arrays.<br>
I would be willing to bet that 80%+ of all the raid arrays out there
are not monitored.<br>
Gluster is more in your face about failures and arguably should be
more reliable in practice because you will know quickly about a
failure.<br>
<br>
Feel free to correct my misconceptions.<br>
<br>
<pre class="moz-signature" cols="72">--
Alvin Starr || voice: (905)513-7688
Netvel Inc. || Cell: (416)806-0133
<a class="moz-txt-link-abbreviated" href="mailto:alvin@netvel.net">alvin@netvel.net</a> ||
</pre>
</body>
</html>