Dynamic memory allocation has been a fundamental part of most computer systems since roughly 1960, and memory allocation is widely considered to be either a solved problem or an insoluble one. In this survey, we describe a variety of memory allocator designs and point out issues relevant to their design and evaluation. We then chronologically survey most of the literature on allocators between 1961 and 1995. (Scores of papers are discussed, in varying detail, and over 150 references are given.)

/pdf/dynamic-storage-allocation-a-survey-and-critical-review-2mi0efvca4.pdf

Dynamic Storage Allocation: A Survey and Critical Review

This paper presents a comprehensive design overview of the SunOS 5.4 kernel memory allocator. This allocator is based on a set of object-caching primitives that reduce the cost of allocating complex objects by retaining their state between uses. These same primitives prove equally effective for managing stateless memory (e.g. data pages and temporary buffers) because they are space-efficient and fast. The allocator's object caches respond dynamically to global memory pressure, and employ an object-coloring scheme that improves the system's overall cache utilization and bus balance. The allocator also has several statistical and debugging features that can detect a wide range of problems throughout the system.

/pdf/the-slab-allocator-an-object-caching-kernel-memory-allocator-kjehvirtct.pdf

The slab allocator: an object-caching kernel memory allocator

Two algorithms are presented for implementing any of a class of buddy systems for dynamic storage allocation. Each buddy system corresponds to a set of recurrence relations which relate the block sizes provided to each other. Analyses of the internal fragmentation of the binary buddy system, the Fibonacci buddy system, and the weighted buddy system are given. Comparative simulation results are also presented for internal, external, and total fragmentation.

Buddy systems

The quantitative evaluation of computer performance is needed during the entire life cycle of a computer system. We survey the major quantitative methods used in computer performance evaluation, focusing on post-1970 developments and emphasizing trends and challenges. We divide the methods used into three main areas, namely performance measurement, analytic performance modeling, and simulation performance modeling, which we survey in the three main sections of the paper. Although we concentrate on the methods per se, rather than on the results of applying the methods, numerous application examples are cited. The methods to be covered have been applied across the entire spectrum of computer systems from personal computers to large mainframes and supercomputers, including both centralized and distributed systems. The application of these methods has certainly not decreased over the years and we anticipate their continued use as well as their enhancement when needed to evaluate future systems.

Computer Performance Evaluation Methodology

The allocation and disposal of memory is a ubiquitous operation in most programs. Rarely do programmers concern themselves with details of memory allocators; most assume that memory allocators provided by the system perform well. Yet, in some applications, programmers use domain-specific knowledge in an attempt to improve the speed or memory utilization of memory allocators.



In this paper, we describe a program (CustoMalloc) that synthesizes a memory allocator customized for a specific application. Our experiments show that the synthesized allocators are uniformly faster and more space efficient than the Berkeley UNIX allocator. Constructing a custom allocator requires little programmer effort, usually taking only a few minutes. Experience has shown that the synthesized allocators are not overly sensitive to properties of input sets and the resulting allocators are superior even to domain-specific allocators designed by programmers. Measurements show that synthesized allocators are from two to ten times faster than widely-used allocators.

CustoMalloc: efficient synthesized memory allocators

Central to successful operation of a computer system based heavily on reentrant (or pure) procedures is reliable and efficient dynamic management of free storage. Such systems must allocate, use, and release one or more blocks of free storage for each task or system operation, e.g., each I/O task or each request for supervisor services. For any list-processing system, such as AED or LISP, efficient management of free storage is a fundamental problem. The consequence of errors in allocation and release is usually total collapse of the system; that of mismanagement is usually processor inefficiency or under-utilization of the free-storage pool. The processor inefficiency resulting from a poor or ill-chosen management algorithm is usually tolerated, even though it is high relative to that of other system functions; under-utilization of free storage is less tolerable, as “lock-up” can be encountered. This circumstance must be averted by task deferral or “garbage collection” procedures, both of which are costly to system performance.

Analysis of free-storage algorithms

Dynamic management of free storage in a time-sharing operating system was studied empirically by the techniques of monitoring, emulation, and on-line experimentation. A new algorithm, based on observed usage patterns of different block sizes, was implemented and evaluated. On-line experiments demonstrated that supervisor time spent in free-storage management was reduced by seven or eight to one.

Barry H. Margolin

Papers

Analysis of free-storage algorithms

Analysis of free-storage algorithms