<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
Well, you could use two file descriptors, one for O_DIRECT one otherwise. <span id="🙂">
🙂</span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
<span><br>
</span></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
SSD is a fast medium but my instinct is the desirability of having data in RAM is much more about I/O pattern and hard to optimize for in advance - Do you read the data you wrote? (Or read data repeatedly?)</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
In any case, there's no mechanism today. It's also relatively marginal if we're just doing buffered I/O then forcing the data out - it will reduce memory usage but it won't improve performance.</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
-Patrick</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);" class="elementToProof">
</div>
<div id="appendonsend"></div>
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
<br>
</div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>From:</b> John Bauer <bauerj@iodoctors.com><br>
<b>Sent:</b> Thursday, May 19, 2022 1:16 PM<br>
<b>To:</b> Patrick Farrell <pfarrell@ddn.com>; lustre-discuss@lists.lustre.org <lustre-discuss@lists.lustre.org><br>
<b>Subject:</b> Re: [lustre-discuss] Avoiding system cache when using ssd pfl extent</font>
<div> </div>
</div>
<div>
<p style="margin-top: 0px; margin-bottom: 0px;">Pat,</p>
<p style="margin-top: 0px; margin-bottom: 0px;">No, not in general. It just seems that if one is storing data on an SSD it should be optional to have it not stored in memory ( why store in 2 fast mediums ).</p>
<p style="margin-top: 0px; margin-bottom: 0px;">O_DIRECT is not of value as that would apply to all extents, whether on SSD on HDD. O_DIRECT on Lustre has been problematic for me in the past, performance wise.<br>
</p>
<p style="margin-top: 0px; margin-bottom: 0px;">John<br>
</p>
<div class="x_moz-cite-prefix">On 5/19/22 13:05, Patrick Farrell wrote:<br>
</div>
<blockquote type="cite">
<div class="x_elementToProof" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
No, and I'm not sure I agree with you at first glance.</div>
<div class="x_elementToProof" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
<br>
</div>
<div class="x_elementToProof" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
Is this just generally an idea that data stored on SSD should not be in RAM? If so, there's no mechanism for that other than using direct I/O.</div>
<div class="x_elementToProof" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
<br>
</div>
<div class="x_elementToProof" style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
-Patrick</div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="x_divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>From:</b> lustre-discuss
<a href="mailto:lustre-discuss-bounces@lists.lustre.org" data-auth="NotApplicable" class="x_moz-txt-link-rfc2396E">
<lustre-discuss-bounces@lists.lustre.org></a> on behalf of John Bauer <a href="mailto:bauerj@iodoctors.com" data-auth="NotApplicable" class="x_moz-txt-link-rfc2396E">
<bauerj@iodoctors.com></a><br>
<b>Sent:</b> Thursday, May 19, 2022 12:48 PM<br>
<b>To:</b> <a href="mailto:lustre-discuss@lists.lustre.org" data-auth="NotApplicable" class="x_moz-txt-link-abbreviated">
lustre-discuss@lists.lustre.org</a> <a href="mailto:lustre-discuss@lists.lustre.org" data-auth="NotApplicable" class="x_moz-txt-link-rfc2396E">
<lustre-discuss@lists.lustre.org></a><br>
<b>Subject:</b> [lustre-discuss] Avoiding system cache when using ssd pfl extent</font>
<div> </div>
</div>
<div class="x_BodyFragment"><font size="2"><span style="font-size:11pt">
<div class="x_PlainText">When using PFL, and using an SSD as the first extent, it seems it would
<br>
be advantageous to not have that extent's file data consume memory in <br>
the client's system buffers. It would be similar to using O_DIRECT, but <br>
on a per-extent basis. Is there a mechanism for that already?<br>
<br>
Thanks,<br>
<br>
John<br>
<br>
_______________________________________________<br>
lustre-discuss mailing list<br>
<a href="mailto:lustre-discuss@lists.lustre.org" data-auth="NotApplicable" class="x_moz-txt-link-abbreviated">lustre-discuss@lists.lustre.org</a><br>
<a href="http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org" data-auth="NotApplicable" class="x_moz-txt-link-freetext">http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org</a><br>
</div>
</span></font></div>
</blockquote>
</div>
</body>
</html>