<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta http-equiv="Content-Language" content="en-us">
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<link href="pool.css" rel="stylesheet" type="text/css">
<title>Pool Concepts</title>
</head>
<body>
<img src="../../../boost.png" width="276" height="86" alt="C++ Boost">
<h1 align="center">Pool Concepts</h1>
<blockquote>
"Dynamic memory allocation has been a fundamental part of most computer
systems since roughly 1960..."<sup><a href="#ref1">1</a></sup>
</blockquote>
<p>Everyone uses dynamic memory allocation. If you have ever called
<span class="code">malloc</span> or <span class="code">new</span>, then you
have used dynamic memory allocation. Most programmers have a tendency to
treat the heap as a "magic bag": we ask it for memory, and it magically
creates some for us. Sometimes we run into problems because the heap is
<em>not</em> magic.</p>
<p>The heap is limited. Even on large systems (i.e., not embedded) with
huge amounts of virtual memory available, there is a limit. Everyone is
aware of the physical limit, but there is a more subtle, "virtual" limit,
that limit at which your program (or the entire system) slows down due to
the use of virtual memory. This virtual limit is much closer to your
program than the physical limit, especially if you are running on a
multitasking system. Therefore, when running on a large system, it is
considered "nice" to make your program use as few resources as necessary,
and release them as soon as possible. When using an embedded system,
programmers usually have no memory to waste.</p>
<p>The heap is complicated. It has to satisfy any type of memory request,
for any size, and do it <em>fast</em>. The common approaches to memory
management have to do with splitting the memory up into portions, and
keeping them ordered by size in some sort of a tree or list structure. Add
in other factors, such as locality and estimating lifetime, and heaps
quickly become very complicated. So complicated, in fact, that there is no
known "perfect" answer to the problem of how to do dynamic memory
allocation. The diagrams below illustrate how most common memory managers
work: for each chunk of memory, it uses part of that memory to maintain its
internal tree or list structure. Even when a chunk is <span class=
"code">malloc</span>'ed out to a program, the memory manager must "save"
some information in it — usually just its size. Then, when the block
is <span class="code">free</span>'d, the memory manager can easily tell how
large it is.</p>
<table cellspacing="0" border="3" rules="none" style=
"float: left; clear: both;" summary="">
<caption>
<em>Memory block, not allocated</em>
</caption>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: silver; text-align: center;">
Memory used internally by memory allocator algorithm (usually 8-12
bytes)</td>
</tr>
<tr>
<td style=
"padding: 2em 0em; background-color: gray; text-align: center">Unused
memory</td>
</tr>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
</table>
<table cellspacing="0" border="3" rules="none" style=
"float: right; clear: both;" summary="">
<caption>
<em>Memory block, allocated (used by program)</em>
</caption>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
<tr>
<td style="background-color: silver; text-align: center;">Memory used
internally by memory allocator algorithm (usually 4 bytes)</td>
</tr>
<tr>
<td style=
"padding: 3em 0em; background-color: yellow; text-align: center">Memory
usable by program</td>
</tr>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
</table>
<p>Because of the complication of dynamic memory allocation, it is often
inefficient in terms of time and/or space. Most memory allocation
algorithms store some form of information with each memory block, either
the block size or some relational information, such as its position in the
internal tree or list structure. It is common for such "header fields" to
take up one machine word in a block that is being used by the program. The
obvious problem, then, is when small objects are dynamically allocated. For
example, if <span class="code">int</span>s were dynamically allocated, then
automatically the algorithm will reserve space for the header fields as
well, and we end up with a 50% waste of memory. Of course, this is a
worst-case scenario. However, more modern programs are making use of small
objects on the heap; and that is making this problem more and more
apparent. Wilson <em>et. al.</em> state that an average-case memory
overhead is about ten to twenty percent<sup><a href="#ref2">2</a></sup>.
This memory overhead will grow higher as more programs use more smaller
objects. It is this memory overhead that brings programs closer to the
virtual limit.</p>
<p>In larger systems, the memory overhead is not as big of a problem
(compared to the amount of time it would take to work around it), and thus
is often ignored. However, there are situations where many allocations
and/or deallocations of smaller objects are taking place as part of a
time-critical algorithm, and in these situations, the system-supplied
memory allocator is often too slow.</p>
<p>Simple segregated storage addresses both of these issues. Almost all
memory overhead is done away with, and all allocations can take place in a
small amount of (amortized) constant time. However, this is done at the
loss of generality; simple segregated storage only can allocate memory
chunks of a single size.</p>
<hr>
<h1 align="center">Simple Segregated Storage</h1>
<p>Simple Segregated Storage is the basic idea behind the Boost Pool
library. Simple Segregated Storage is the simplest, and probably the
fastest, memory allocation/deallocation algorithm. It begins by
<em>partitioning</em> a memory <em>block</em> into fixed-size
<em>chunks</em>. Where the block comes from is not important until
implementation time. A <em>Pool</em> is some object that uses Simple
Segregated Storage in this fashion. To illustrate:</p>
<table cellspacing="0" border="3" rules="none" align="center" style=
"clear: both;" summary="">
<caption>
<em>Memory block, split into chunks</em>
</caption>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
0</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
1</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
2</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
3</td>
</tr>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
</table>
<p>Each of the chunks in any given block are <strong>always</strong> the
same size. This is the fundamental restriction of Simple Segregated
Storage: you cannot ask for chunks of different sizes. For example, you
cannot ask a Pool of integers for a character, or a Pool of characters for
an integer (assuming that characters and integers are different sizes).</p>
<p>Simple Segregated Storage works by interleaving a <em>free list</em>
within the unused chunks. For example:</p>
<table cellspacing="0" border="3" rules="none" style=
"float: left; clear: both;" summary="">
<caption>
<em>Memory block, with no chunks allocated</em>
</caption>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
0; points to Chunk 1</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
1; points to Chunk 2</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
2; points to Chunk 3</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
3; end-of-list</td>
</tr>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
</table>
<table cellspacing="0" border="3" rules="none" style=
"float: right; clear: both;" summary="">
<caption>
<em>Memory block, with two chunks allocated</em>
</caption>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
0; points to Chunk 2</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: silver; text-align: center;">Chunk
1 (in use by process)</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: gray; text-align: center;">Chunk
2; end-of-list</td>
</tr>
<tr>
<td style=
"padding: 1em 0em; background-color: silver; text-align: center;">Chunk
3 (in use by process)</td>
</tr>
<tr>
<td style="background-color: red; text-align: center;">Memory not
belonging to process</td>
</tr>
</table>
<p>By interleaving the free list inside the chunks, each Simple Segregated
Storage only has the overhead of a single pointer (the pointer to the first
element in the list). It has <em>no</em> memory overhead for chunks that
are in use by the process.</p>
<p>Simple Segregated Storage is also extremely fast. In the simplest case,
memory allocation is merely removing the first chunk from the free list, a
O(1) operation. In the case where the free list is empty, another block may
have to be acquired and partitioned, which would result in an amortized
O(1) time. Memory deallocation may be as simple as adding that chunk to the
front of the free list, a O(1) operation. However, more complicated uses of
Simple Segregated Storage may require a sorted free list, which makes
deallocation O(N).</p>
<p>Simple Segregated Storage gives faster execution and less memory
overhead than a system-supplied allocator, but at the loss of generality. A
good place to use a Pool is in situations where many (noncontiguous) small
objects may be allocated on the heap, or if allocation and deallocation of
the same-sized objects happens repeatedly.<br clear="all"></p>
<hr>
<h2>References</h2>
<ol>
<li><a name="ref1" id="ref1">Doug Lea, <em>A Memory Allocator</em>.</a>
Available on the web at <a href=
"http://gee.cs.oswego.edu/dl/html/malloc.html">http://gee.cs.oswego.edu/dl/html/malloc.html</a></li>
<li><a name="ref2" id="ref2">Paul R. Wilson, Mark S. Johnstone, Michael
Neely, and David Boles, "Dynamic Storage Allocation: A Survey and
Critical Review" in <em>International Workshop on Memory Management</em>,
September 1995, pg. 28, 36.</a> Available on the web at <a href=
"ftp://ftp.cs.utexas.edu/pub/garbage/allocsrv.ps">ftp://ftp.cs.utexas.edu/pub/garbage/allocsrv.ps</a></li>
</ol>
<h2>Other Implementations</h2>
<p>Pool allocators are found in many programming languages, and in many
variations. The beginnings of many implementations may be found in common
programming literature; some of these are given below. Note that none of
these are complete implementations of a Pool; most of these leave some
aspects of a Pool as a user exercise. However, in each case, even though
some aspects are missing, these examples use the same underlying concept of
a Simple Segregated Storage described in this document.</p>
<ul>
<li>"The C++ Programming Language", 3rd ed., by Bjarne Stroustrup,
Section 19.4.2. Missing aspects:
<ol>
<li>Not portable</li>
<li>Cannot handle allocations of arbitrary numbers of objects (this
was left as an exercise)</li>
<li>Not thread-safe</li>
<li>Suffers from the static initialization problem</li>
</ol>
</li>
<li>"MicroC/OS-II: The Real-Time Kernel", by Jean J. Labrosse, Chapter 7
and Appendix B.04. This is an example of the Simple Segregated Storage
scheme at work in the internals of an actual OS. Missing aspects:
<ol>
<li>Not portable (though this is OK, since it's part of its own
OS)</li>
<li>Cannot handle allocations of arbitrary numbers of blocks (which
is also OK, since this feature is not needed)</li>
<li>Requires non-intuitive user code to create and destroy the
Pool</li>
</ol>
</li>
<li>"Efficient C++: Performance Programming Techniques", by Dov Bulka and
David Mayhew, Chapters 6 and 7. This is a good example of iteratively
developing a Pool solution; however, their premise (that the
system-supplied allocation mechanism is hopelessly inefficient) is flawed
on every system I've tested on. Run their timings on your system before
you accept their conclusions. Missing aspects:
<ol>
<li>Requires non-intuitive user code to create and destroy the
Pool</li>
</ol>
</li>
<li>"Advanced C++: Programming Styles and Idioms", by James O. Coplien,
Section 3.6. This has examples of both static and dynamic pooling.
Missing aspects:
<ol>
<li>Not thread-safe</li>
<li>The static pooling example is not portable</li>
</ol>
</li>
</ul>
<hr>
<p><a href="http://validator.w3.org/check?uri=referer"><img border="0" src=
"../../../doc/images/valid-html401.png" alt="Valid HTML 4.01 Transitional"
height="31" width="88"></a></p>
<p>Revised
<!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B, %Y" startspan -->05
December, 2006<!--webbot bot="Timestamp" endspan i-checksum="38516" --></p>
<p><i>Copyright © 2000, 2001 Stephen Cleary (scleary AT jerviswebb DOT
com)</i></p>
<p><i>Distributed under the Boost Software License, Version 1.0. (See
accompanying file <a href="../../../LICENSE_1_0.txt">LICENSE_1_0.txt</a> or
copy at <a href=
"http://www.boost.org/LICENSE_1_0.txt">http://www.boost.org/LICENSE_1_0.txt</a>)</i></p>
</body>
</html>
