Difference between revisions of "Address"
Chris Tyler (talk | contribs) (Created page with 'Category:Computer Architecture{{Chris Tyler Draft}} An '''address''' is the numeric identifier for a memory location. The term may also be used for other storage allocation i…') |
Chris Tyler (talk | contribs) |
||
Line 1: | Line 1: | ||
− | [[Category:Computer Architecture]] | + | [[Category:Computer Architecture]] |
An '''address''' is the numeric identifier for a memory location. The term may also be used for other storage allocation identifiers -- for example, a '''logical block address''' on a disk identifies a block or sector on that disk using a unique number. | An '''address''' is the numeric identifier for a memory location. The term may also be used for other storage allocation identifiers -- for example, a '''logical block address''' on a disk identifies a block or sector on that disk using a unique number. | ||
Revision as of 15:14, 6 January 2014
An address is the numeric identifier for a memory location. The term may also be used for other storage allocation identifiers -- for example, a logical block address on a disk identifies a block or sector on that disk using a unique number.
Each memory location is identified by a unique numeric address. The maximum number of unique memory locations is dictated by the width of the address in bits; for example, a 32-bit address can refer to any of 232 addresses, limiting memory to 4GB.
Most modern CPUs with a memory management unit (MMU) separate physical memory addresses from virtual memory addresses -- in other words, they can renumber memory. This feature is usually used on a per-process basis by the operating system, to present a different view of memory to each process.
It is not necessary to fully populate the address space -- a computer with a 32-bit address bus may only have 1GB of installed memory. Reading unpopulated memory locations will typically lead to undefined results.