ibm-information-center/dist/eclipse/plugins/i5OS.ic.dbmult_5.4.0.1/summary.htm

67 lines
4.1 KiB
HTML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html lang="en-us" xml:lang="en-us">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="security" content="public" />
<meta name="Robots" content="index,follow" />
<meta http-equiv="PICS-Label" content='(PICS-1.1 "http://www.icra.org/ratingsv02.html" l gen true r (cz 1 lz 1 nz 1 oz 1 vz 1) "http://www.rsac.org/ratingsv01.html" l gen true r (n 0 s 0 v 0 l 0) "http://www.classify.org/safesurf/" l gen true r (SS~~000 1))' />
<meta name="DC.Type" content="task" />
<meta name="DC.Title" content="Summary of performance considerations" />
<meta name="abstract" content="There are performance factors you should consider when developing queries that use distributed files." />
<meta name="description" content="There are performance factors you should consider when developing queries that use distributed files." />
<meta name="DC.subject" content="query, performance consideration summary, performance, summary for queries" />
<meta name="keywords" content="query, performance consideration summary, performance, summary for queries" />
<meta name="DC.Relation" scheme="URI" content="qqp.htm" />
<meta name="copyright" content="(C) Copyright IBM Corporation 1998, 2006" />
<meta name="DC.Rights.Owner" content="(C) Copyright IBM Corporation 1998, 2006" />
<meta name="DC.Format" content="XHTML" />
<meta name="DC.Identifier" content="summary" />
<meta name="DC.Language" content="en-us" />
<!-- All rights reserved. Licensed Materials Property of IBM -->
<!-- US Government Users Restricted Rights -->
<!-- Use, duplication or disclosure restricted by -->
<!-- GSA ADP Schedule Contract with IBM Corp. -->
<link rel="stylesheet" type="text/css" href="./ibmdita.css" />
<link rel="stylesheet" type="text/css" href="./ic.css" />
<title>Summary of performance considerations</title>
</head>
<body id="summary"><a name="summary"><!-- --></a>
<!-- Java sync-link --><script language="Javascript" src="../rzahg/synch.js" type="text/javascript"></script>
<h1 class="topictitle1">Summary of performance considerations</h1>
<div><p>There are performance factors you should consider when developing
queries that use distributed files.</p>
<ol><li class="stepexpand"><span>For the OPNQRYF command and the query API (QQQQRY), specifying
ALWCPYDTA(*OPTIMIZE) allows each node to choose an index or a sort to satisfy
the ordering specified.</span></li>
<li class="stepexpand"><span>For the OPNQRYF command and the query API (QQQQRY), specifying
ALWCPYDTA(*YES) or ALWCPYDTA(*NO) enforces that each node use an index that
exactly matches the specified ordering fields. This is more restrictive than
the way the optimizer processes ordering for nondistributed files.</span></li>
<li class="stepexpand"><span>Adding an ORDER BY clause to a DISTINCT select can return records
faster by not requiring a final sort on the requesting system.</span></li>
<li class="stepexpand"><span>Including all of the fields of the partitioning key in the grouping
fields generally results in one-step grouping, which performs better than
two-step grouping.</span></li>
<li class="stepexpand"><span>Including all of the fields of the partitioning key in the join
criteria generally results in a collocated distributed join.</span></li>
<li class="stepexpand"><span>Including all of the fields of the partitioning key in isolatable,
equal record selection generally results in the query being processed on only
one node.</span></li>
<li class="stepexpand"><span>Including any of the following scalar functions in isolatable,
equal record selection generally results in the query being processed on only
one node:</span> <ul><li>NODENAME</li>
<li>NODENUMBER</li>
<li>PARTITION</li>
</ul>
</li>
</ol>
</div>
<div>
<div class="familylinks">
<div class="parentlink"><strong>Parent topic:</strong> <a href="qqp.htm" title="This topic provides you with some guidelines for designing queries so that they use query resources more efficiently when you run queries that use distributed files.">Query design for performance with DB2 Multisystem</a></div>
</div>
</div>
</body>
</html>