In computer science an array is a data structure consisting of a group of elements that are accessed by indexing. In most programming languages each element has the same data type and the array occupies a contiguous area of storage.
Some programming languages support array programming (e.g., APL, newer versions of Fortran) which generalises operations and functions to work transparently over arrays as they do with scalars, instead of requiring looping over array members.
Multi-dimensional arrays are accessed using more than one index: one for each dimension. Multidimensional indexing reduced to a lesser number of dimensions, for example, a two-dimensional array with consisting of 6 and 5 elements respectively could be represented using a one-dimensional array of 30 elements.
Arrays can be classified as fixed-sized arrays (sometimes known as static arrays) whose size cannot change once their storage has been allocated, or dynamic arrays, which can be resized.
Memory-wise, arrays are compact data structures with no per-element overhead. There may be a per-array overhead, e.g. to store index bounds, but this is language-dependent. It can also happen that elements stored in an array require less memory than the same elements stored in individual variables, because several array elements can be stored in a single word; such arrays are often called packed arrays.
Associative arrays provide a mechanism for array-like functionality without huge storage overheads when the index values are sparse. Specialized associative arrays with integer keys include Patricia tries and Judy arrays.
Balanced trees require O(log n) time for index access, but also permit inserting or deleting elements in Θ(log n) time. Arrays require O(n) time for insertion and deletion of elements.
One or more large arrays are sometimes used to emulate in-program dynamic memory allocation, particularly memory pool allocation. Historically, this has sometimes been the only way to allocate "dynamic memory" portably.
Array accesses with statically predictable access patterns are a major source of data parallelism.
The valid index values of each dimension of an array are a bounded set of integers. Programming environments that check indexes for validity are said to perform bounds checking.
The Comparison of programming languages (array), indicates the base index used by various languages.
Supporters of zero-based indexing sometimes criticize one-based and n-based arrays for being slower. Often this criticism is mistaken when one-based or n-based array accesses are optimized with common subexpression elimination (for single dimensioned arrays) and/or with well-defined dope vectors (for multi-dimensioned arrays). However, in multidimensional arrays where the net offset into linear memory is computed from all of the indices, zero-based indexing is more natural, simpler, and faster. Edsger W. Dijkstra expressed an opinion in this debate: Why numbering should start at zero
The 0-based/1-based debate is not limited to just programming languages. For example, the ground-floor of a building is elevator button "0" in France, but elevator button "1" in the USA.
Common ways to index into multi-dimensional arrays include:
The first two forms are more compact and have potentially better locality of reference, but are also more limiting; the arrays must be rectangular, meaning that no row can contain more elements than any other. Arrays of arrays, on the other hand, allow the creation of ragged arrays, also called jagged arrays, in which the valid range of one index depends on the value of another, or in this case, simply that different rows can be different sizes. Arrays of arrays are also of value in programming languages that only supply one-dimensional arrays as primitives.
In many applications, such as numerical applications working with matrices, we iterate over rectangular two-dimensional arrays in predictable ways. For example, computing an element of the matrix product AB involves iterating over a row of A and a column of B simultaneously. In mapping the individual array indexes into memory, we wish to exploit locality of reference as much as we can. A compiler can sometimes automatically choose the layout for an array so that sequentially accessed elements are stored sequentially in memory; in our example, it might choose row-major order for A, and column-major order for B. Even more exotic orderings can be used, for example if we iterate over the main diagonal of a matrix.
Cache design. (includes related articles on reducing disk-access times with caches and reducing pipeline stalls ) (Cover Story)
Apr 25, 1991; Deciding how to implement a cache is like buying a car: after you decide which car, you have to choose the color, leather or...
Cache memories and MMUs work together in high-speed systems. (includes a related article on basic cache considerations) (tutorial)
Mar 15, 1990; Cache memories and MMUs work together in high-speed systems Placing a static-RAM (SRAM) cache-memory system between a CPU and the...