관리-도구
편집 파일: performance.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en"> <head> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /> <meta name="generator" content="AsciiDoc 8.5.3" /> <title>Dstat performance</title> <style type="text/css"> /* Debug borders */ p, li, dt, dd, div, pre, h1, h2, h3, h4, h5, h6 { /* border: 1px solid red; */ } body { margin: 1em 5% 1em 5%; } a { color: blue; text-decoration: underline; } a:visited { color: fuchsia; } em { font-style: italic; color: navy; } strong { font-weight: bold; color: #083194; } tt { color: navy; } h1, h2, h3, h4, h5, h6 { color: #527bbd; font-family: sans-serif; margin-top: 1.2em; margin-bottom: 0.5em; line-height: 1.3; } h1, h2, h3 { border-bottom: 2px solid silver; } h2 { padding-top: 0.5em; } h3 { float: left; } h3 + * { clear: left; } div.sectionbody { font-family: serif; margin-left: 0; } hr { border: 1px solid silver; } p { margin-top: 0.5em; margin-bottom: 0.5em; } ul, ol, li > p { margin-top: 0; } pre { padding: 0; margin: 0; } span#author { color: #527bbd; font-family: sans-serif; font-weight: bold; font-size: 1.1em; } span#email { } span#revnumber, span#revdate, span#revremark { font-family: sans-serif; } div#footer { font-family: sans-serif; font-size: small; border-top: 2px solid silver; padding-top: 0.5em; margin-top: 4.0em; } div#footer-text { float: left; padding-bottom: 0.5em; } div#footer-badges { float: right; padding-bottom: 0.5em; } div#preamble { margin-top: 1.5em; margin-bottom: 1.5em; } div.tableblock, div.imageblock, div.exampleblock, div.verseblock, div.quoteblock, div.literalblock, div.listingblock, div.sidebarblock, div.admonitionblock { margin-top: 1.0em; margin-bottom: 1.5em; } div.admonitionblock { margin-top: 2.0em; margin-bottom: 2.0em; margin-right: 10%; color: #606060; } div.content { /* Block element content. */ padding: 0; } /* Block element titles. */ div.title, caption.title { color: #527bbd; font-family: sans-serif; font-weight: bold; text-align: left; margin-top: 1.0em; margin-bottom: 0.5em; } div.title + * { margin-top: 0; } td div.title:first-child { margin-top: 0.0em; } div.content div.title:first-child { margin-top: 0.0em; } div.content + div.title { margin-top: 0.0em; } div.sidebarblock > div.content { background: #ffffee; border: 1px solid silver; padding: 0.5em; } div.listingblock > div.content { border: 1px solid silver; background: #f4f4f4; padding: 0.5em; } div.quoteblock, div.verseblock { padding-left: 1.0em; margin-left: 1.0em; margin-right: 10%; border-left: 5px solid #dddddd; color: #777777; } div.quoteblock > div.attribution { padding-top: 0.5em; text-align: right; } div.verseblock > pre.content { font-family: inherit; } div.verseblock > div.attribution { padding-top: 0.75em; text-align: left; } /* DEPRECATED: Pre version 8.2.7 verse style literal block. */ div.verseblock + div.attribution { text-align: left; } div.admonitionblock .icon { vertical-align: top; font-size: 1.1em; font-weight: bold; text-decoration: underline; color: #527bbd; padding-right: 0.5em; } div.admonitionblock td.content { padding-left: 0.5em; border-left: 3px solid #dddddd; } div.exampleblock > div.content { border-left: 3px solid #dddddd; padding-left: 0.5em; } div.imageblock div.content { padding-left: 0; } span.image img { border-style: none; } a.image:visited { color: white; } dl { margin-top: 0.8em; margin-bottom: 0.8em; } dt { margin-top: 0.5em; margin-bottom: 0; font-style: normal; color: navy; } dd > *:first-child { margin-top: 0.1em; } ul, ol { list-style-position: outside; } ol.arabic { list-style-type: decimal; } ol.loweralpha { list-style-type: lower-alpha; } ol.upperalpha { list-style-type: upper-alpha; } ol.lowerroman { list-style-type: lower-roman; } ol.upperroman { list-style-type: upper-roman; } div.compact ul, div.compact ol, div.compact p, div.compact p, div.compact div, div.compact div { margin-top: 0.1em; margin-bottom: 0.1em; } div.tableblock > table { border: 3px solid #527bbd; } thead, p.table.header { font-family: sans-serif; font-weight: bold; } tfoot { font-weight: bold; } td > div.verse { white-space: pre; } p.table { margin-top: 0; } /* Because the table frame attribute is overriden by CSS in most browsers. */ div.tableblock > table[frame="void"] { border-style: none; } div.tableblock > table[frame="hsides"] { border-left-style: none; border-right-style: none; } div.tableblock > table[frame="vsides"] { border-top-style: none; border-bottom-style: none; } div.hdlist { margin-top: 0.8em; margin-bottom: 0.8em; } div.hdlist tr { padding-bottom: 15px; } dt.hdlist1.strong, td.hdlist1.strong { font-weight: bold; } td.hdlist1 { vertical-align: top; font-style: normal; padding-right: 0.8em; color: navy; } td.hdlist2 { vertical-align: top; } div.hdlist.compact tr { margin: 0; padding-bottom: 0; } .comment { background: yellow; } .footnote, .footnoteref { font-size: 0.8em; } span.footnote, span.footnoteref { vertical-align: super; } #footnotes { margin: 20px 0 20px 0; padding: 7px 0 0 0; } #footnotes div.footnote { margin: 0 0 5px 0; } #footnotes hr { border: none; border-top: 1px solid silver; height: 1px; text-align: left; margin-left: 0; width: 20%; min-width: 100px; } @media print { div#footer-badges { display: none; } } div#toc { margin-bottom: 2.5em; } div#toctitle { color: #527bbd; font-family: sans-serif; font-size: 1.1em; font-weight: bold; margin-top: 1.0em; margin-bottom: 0.1em; } div.toclevel1, div.toclevel2, div.toclevel3, div.toclevel4 { margin-top: 0; margin-bottom: 0; } div.toclevel2 { margin-left: 2em; font-size: 0.9em; } div.toclevel3 { margin-left: 4em; font-size: 0.9em; } div.toclevel4 { margin-left: 6em; font-size: 0.9em; } /* Workarounds for IE6's broken and incomplete CSS2. */ div.sidebar-content { background: #ffffee; border: 1px solid silver; padding: 0.5em; } div.sidebar-title, div.image-title { color: #527bbd; font-family: sans-serif; font-weight: bold; margin-top: 0.0em; margin-bottom: 0.5em; } div.listingblock div.content { border: 1px solid silver; background: #f4f4f4; padding: 0.5em; } div.quoteblock-attribution { padding-top: 0.5em; text-align: right; } pre.verseblock-content { font-family: inherit; } div.verseblock-attribution { padding-top: 0.75em; text-align: left; } div.exampleblock-content { border-left: 3px solid #dddddd; padding-left: 0.5em; } /* IE6 sets dynamically generated links as visited. */ div#toc a:visited { color: blue; } </style> <script type="text/javascript"> /*<+'])'); // Function that scans the DOM tree for header elements (the DOM2 // nodeIterator API would be a better technique but not supported by all // browsers). var iterate = function (el) { for (var i = el.firstChild; i != null; i = i.nextSibling) { if (i.nodeType == 1 /* Node.ELEMENT_NODE */) { var mo = re.exec(i.tagName); if (mo && (i.getAttribute("class") || i.getAttribute("className")) != "float") { result[result.length] = new TocEntry(i, getText(i), mo[1]-1); } iterate(i); } } } iterate(el); return result; } var toc = document.getElementById("toc"); var entries = tocEntries(document.getElementById("content"), toclevels); for (var i = 0; i < entries.length; ++i) { var entry = entries[i]; if (entry.element.id == "") entry.element.id = "_toc_" + i; var a = document.createElement("a"); a.href = "#" + entry.element.id; a.appendChild(document.createTextNode(entry.text)); var div = document.createElement("div"); div.appendChild(a); div.className = "toclevel" + entry.toclevel; toc.appendChild(div); } if (entries.length == 0) toc.parentNode.removeChild(toc); }, ///////////////////////////////////////////////////////////////////// // Footnotes generator ///////////////////////////////////////////////////////////////////// /* Based on footnote generation code from: * http://www.brandspankingnew.net/archive/2005/07/format_footnote.html */ footnotes: function () { var cont = document.getElementById("content"); var noteholder = document.getElementById("footnotes"); var spans = cont.getElementsByTagName("span"); var refs = {}; var n = 0; for (i=0; i<spans.length; i++) { if (spans[i].className == "footnote") { n++; // Use [\s\S] in place of . so multi-line matches work. // Because JavaScript has no s (dotall) regex flag. note = spans[i].innerHTML.match(/\s*\[([\s\S]*)]\s*/)[1]; noteholder.innerHTML += "<div class='footnote' id='_footnote_" + n + "'>" + "<a href='#_footnoteref_" + n + "' title='Return to text'>" + n + "</a>. " + note + "</div>"; spans[i].innerHTML = "[<a id='_footnoteref_" + n + "' href='#_footnote_" + n + "' title='View footnote' class='footnote'>" + n + "</a>]"; var id =spans[i].getAttribute("id"); if (id != null) refs["#"+id] = n; } } if (n == 0) noteholder.parentNode.removeChild(noteholder); else { // Process footnoterefs. for (i=0; i<spans.length; i++) { if (spans[i].className == "footnoteref") { var href = spans[i].getElementsByTagName("a")[0].getAttribute("href"); href = href.match(/#.*/)[0]; // Because IE return full URL. n = refs[href]; spans[i].innerHTML = "[<a href='#_footnote_" + n + "' title='View footnote' class='footnote'>" + n + "</a>]"; } } } } } /*]]>*/ </script> </head> <body> <div id="header"> <h1>Dstat performance</h1> </div> <div id="content"> <h2 id="_introduction">Introduction</h2> <div class="sectionbody"> <div class="paragraph"><p>Since Dstat is written in python, it is not optimized for performance. But that doesn’t mean that Dstat performs bad, it performs quite good given its written in python and a lot of dedication went into profiling and optimizing Dstat and Dstat plugins.</p></div> <div class="paragraph"><p>But when doing performance analysis, it is always important to verify that the monitoring tool is not interfering with the performance numbers. (eg. writing to disk, using cpu/memory/network, increasing load)</p></div> </div> <h2 id="_compare_with_baseline">Compare with baseline</h2> <div class="sectionbody"> <div class="paragraph"><p>Depending on the plugins being used and the load on the server itself the impact Dstat has on the system you are monitoring might be considerable. A lot of plugins are pretty fast (less than 0.1ms on an modest 1.2Ghz laptop), but some plugins may use up to 3ms or even up to 2% of your CPU. (eg. each top-plugin scans the process-list)</p></div> <div class="paragraph"><p>Before performing any tests please verify for yourself what impact Dstat has on your test results and keep that in mind when analysing the results afterwards. Especially if you suspect Dstat to be influencing your results, do a baseline with and without the Dstat commandline.</p></div> </div> <h2 id="_selection_of_plugins">Selection of plugins</h2> <div class="sectionbody"> <div class="paragraph"><p>In case the impact is higher than expected, reduce the number of plugins and remove expensive plugins, or even better, look at the plugin you’re using and send me optimizations.</p></div> <div class="paragraph"><p>Newer python versions are also faster than older ones, and hardware is only becoming faster at a pace that these considerations may not hold anylonger.</p></div> </div> <h2 id="_debugging_and_profiling_dstat">Debugging and profiling Dstat</h2> <div class="sectionbody"> <div class="paragraph"><p>If you need feedback about plugin performance, use the --debug option to profile different plugins. If you use -t together with --debug, you can see the time deviation on your system in relation to load/plugins.</p></div> <div class="paragraph"><p>If you want to profile certain plugins, you can use the --profile option which provides you with detailed information of the function calls that are the most expensive.</p></div> <div class="paragraph"><p>You can also run the dstat plugin (--dstat) to look what overhead (cputime) and response (latency) Dstat has during runtime, which can be very useful to compare with your baseline and the system in idle state.</p></div> <div class="paragraph"><p>One common way to profile a single plugin is to use the following commandline:</p></div> <div class="literalblock"> <div class="content"> <pre><tt>dstat -t --dstat --debug --profile dstat -t --dstat --top-cpu --debug --profile</tt></pre> </div></div> <div class="paragraph"><p>The default profiling infrastructure is quite expensive, so it is important that you first make a baseline including the profiling itself, then compare it against the same commandline including the plugin you want to profile.</p></div> </div> <h2 id="_improving_dstat_8217_s_footprint_even_more">Improving Dstat’s footprint even more</h2> <div class="sectionbody"> <div class="paragraph"><p>Another way to win a few CPU cycles is to pre-compile the Dstat plugins by running the compileall.py script that comes with python on your plugins directory. It can save about 10% in execution time.</p></div> <div class="paragraph"><p>Remember that invisible plugins (that run out of your terminal window) do take up cycles because the information is still being collected and possibly written to CSV output.</p></div> <div class="paragraph"><p>It should be possible to write plugins in C to improve the impact on the system, but I have no experience with writing python modules in C. Any feedback on this is welcomed.</p></div> </div> <h2 id="_performance_tuning">Performance tuning</h2> <div class="sectionbody"> <div class="paragraph"><p>The following documents may be useful to tune a system for performance</p></div> <div class="ulist"><ul> <li> <p> <a href="http://people.redhat.com/alikins/system_tuning.html">http://people.redhat.com/alikins/system_tuning.html</a> </p> </li> </ul></div> <div class="admonitionblock"> <table><tr> <td class="icon"> <div class="title">Note</div> </td> <td class="content">Please send me improvements to this document.</td> </tr></table> </div> </div> </div> <div id="footnotes"><hr /></div> <div id="footer"> <div id="footer-text"> Last updated 2010-02-11 11:24:41 CEST </div> </div> </body> </html>