The first time you sign into developerWorks, a profile is created for you. Select information in your profile (name, country/region, and company) is displayed to the public and will accompany any content you post. You may update your IBM account at any time.

All information submitted is secure.

The first time you sign in to developerWorks, a profile is created for you, so you need to choose a display name. Your display name accompanies the content you post on developerworks.

Please choose a display name between 3-31 characters. Your display name must be unique in the developerWorks community and should not be your email address for privacy reasons.

All information submitted is secure.

My developerWorks:

My notifications:

Sign out

Select a language:

Building your own memory manager for C/C++ projects

Arpan Sen (arpan@syncad.com), Technical lead, Synapti Computer Aided Design Pvt Ltd

Arpan Sen is a lead engineer working on the development of software in the electronic design automation industry. He has worked on several flavors of UNIX, including Solaris, SunOS, HP-UX, and IRIX as well as Linux and Microsoft Windows for several years. He takes a keen interest in software performance-optimization techniques, graph theory, and parallel computing. Arpan holds a post-graduate degree in software systems. You can reach him at arpansen@gmail.com.
(An IBM developerWorks Contributing Author)

Rahul Kumar Kardam (rahul@syncad.com), Senior Software Developer, Synapti Computer Aided Design Pvt Ltd

Rahul is a senior software developer specializing in developing complex C++-based electronic design automation tools such as simulators for hardware designs. He has programming experience in both Windows and UNIX platforms. Rahul enjoys tinkering with open source software, which he uses as a framework for designing robust, scalable code for the design automation tools he works on. He also enjoys technical writing.

Summary: Performance optimization of code is serious business. It's fairly common to see a piece of functionally correct software written in C or C++ that takes way too much memory, time, or, in the worst case, both. As a developer, one of the most powerful tools that C/C++ arms you with to improve processing time and prevent memory corruption is the control over how memory is allocated or deallocated in your code. This tutorial demystifies memory management concepts by telling you how to create your very own memory manager for specific situations.

Date: 19 Feb 2008
Level: Intermediate
PDF: A4 and Letter (83 KB)Get Adobe® Reader®

Activity: 70491 views
Comments:

Why create a custom memory manager?

To appreciate how the control over memory allocation helps your code run faster, first recollect the basics of memory management in C/C++. The standard library functions malloc, free, calloc, and realloc in C and the new, new [ ], delete, and delete [ ] operators in C++ form the crux of the memory management in these two languages. With this in mind, there are several things worth your attention here.

Functions such as malloc and new are general-purpose memory allocators. Your code may be single-threaded, but the malloc function it is linked to can handle multithreaded paradigms just as well. It is this extra functionality that degrades the performance of these routines.

In their turn, malloc and new make calls to the operating system kernel requesting memory, while free and delete make requests to release memory. This means that the operating system has to switch between user-space code and kernel code every time a request for memory is made. Programs making repeated calls to malloc or new eventually run slowly because of the repeated context switching.

Memory that is allocated in a program and subsequently not needed is often unintentionally left undeleted, and C/C++ don't provide for automatic garbage collection. This causes the memory footprint of the program to increase. In the case of really big programs, performance takes a severe hit because available memory becomes increasingly scarce and hard-disk accesses are time intensive.

Comments