📄 lib0040.html

📁 Memory Management—Algorithms and implementation in C/C++ Introduction Chapter 1 - Memory Manag
💻 HTML
字号:
<html>
<META http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<title>Performance Comparison</title>
<link rel="STYLESHEET" type="text/css" href="images/xpolecat.css">
<link rel="STYLESHEET" type="text/css" href="images/ie.content.books24x7.css">
</head>
<body >
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<td><div STYLE="MARGIN-LEFT: 0.15in;">
<a href="toc.html"><img src="images/teamlib.gif" width="62" height="15" border="0" align="absmiddle"  alt="Team LiB"></a></div></td>
<td valign="top" class="v2" align="right"><div STYLE="MARGIN-RIGHT: 0.15in"><a href="LiB0039.html"><img src="images/previous.gif" width="62" height="15" border="0" align="absmiddle" alt="Previous Section"></a>
<a href="LiB0041.html"><img src="images/next.gif" width="41" height="15" border="0" align="absmiddle" alt="Next Section"></a>
</div></td></tr>
</table>

<div class="chapter">
<a name="ch04"></a>
<div class="section">
<h2 class="first-section-title"><a name="531"></a><a name="ch04lev1sec9"></a>Performance Comparison</h2><p class="first-para">Let us revisit all three of the previous implementations. In each of the performance tests, the resident memory manager was given 1MB of storage and was asked to service 1,024 consecutive memory allocation/release operations. The score card is provided in <a class="internaljump" href="#ch04table05">Table 4.5</a>.</p>
<a name="532"></a><a name="ch04table05"></a>
<table class="table" border="1">
<caption class="table-title">
<span class="table-title"><span class="table-titlelabel">Table 4.5</span></span>
</caption>
<thead>
<tr valign="top">
<th class="th" scope="col" align="left">
<p class="table-para">
<b class="bold">Manager</b>
</p>
</th><th class="th" scope="col" align="left">
<p class="table-para">
<b class="bold">Milliseconds</b>
</p>
</th>
</tr>
</thead>
<tbody>
<tr valign="top">
<td class="td" align="left">
<p class="table-para">bitmapped</p>
</td><td class="td" align="left">
<p class="table-para">856</p>
</td>
</tr>
<tr valign="top">
<td class="td" align="left">
<p class="table-para">sequential fit</p>
</td><td class="td" align="left">
<p class="table-para">35</p>
</td>
</tr>
<tr valign="top">
<td class="td" align="left">
<p class="table-para">segregated list</p>
</td><td class="td" align="left">
<p class="table-para">5</p>
</td>
</tr>
</tbody>
</table>
<p class="para">By a long shot, the bitmapped memory manager was the worst. Now you know why almost nobody uses it. The segregated memory manager outperformed the sequential fit memory manager. However, there are significant trade-offs that might make the sequential fit method preferable to the segregated list method.</p>
<p class="para">One common theme you will see in your journey through computer science is the <i class="emphasis">storage-versus-speed</i> trade-off. If you increase the amount of available memory storage, you can usually make something run faster.</p>
<p class="para">Here is an example: If you rewrite all of the functions in a program as inline macros, you can increase the speed at which the program executes. This is because the processor doesn't have to <a name="533"></a><a name="IDX-280"></a>waste time jumping around to different memory locations. In addition, because execution will not frequently jump to a nonlocal spot in memory, the processor will be able to spend much of its time executing code in the cache.</p>
<p class="para">Likewise, you can make a program smaller by isolating every bit of redundant code and placing it in its own function. While this will decrease the total number of machine instructions, this tactic will make a program slower because not only does the processor spend most of its time jumping around memory, but the processor's cache will also need to be constantly refreshed.</p>
<p class="last-para">This is the type of situation that we face with the segregated list and sequential fit approaches. The segregated list approach is much faster, but it also wastes a lot of memory. The sequential fit algorithm, on the other hand, decreases the amount of wasted storage at the expense of execution time. This leads me to think that the segregated list approach would be useful for embedded systems that do not typically have much memory to spend. The segregated storage approach might be suited for large enterprise servers that are running high-performance transaction processing software.</p>
<a></a>
</div>
</div>
</div>
</div>
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<td><div STYLE="MARGIN-LEFT: 0.15in;">
<a href="toc.html"><img src="images/teamlib.gif" width="62" height="15" border="0" align="absmiddle"  alt="Team LiB"></a></div></td>
<td valign="top" class="v2" align="right"><div STYLE="MARGIN-RIGHT: 0.15in"><a href="LiB0039.html"><img src="images/previous.gif" width="62" height="15" border="0" align="absmiddle" alt="Previous Section"></a>
<a href="LiB0041.html"><img src="images/next.gif" width="41" height="15" border="0" align="absmiddle" alt="Next Section"></a>
</div></td></tr>
</table>
</body>
</html>
⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -