Shared Memory Architecture

Shared Memory Architecture

In computer architecture, Shared Memory Architecture (SMA) refers to a design where the graphics chip does not have its own dedicated memory, and instead shares the main system RAM with the CPU and other components.

This design is used with many integrated graphics solutions to reduce the cost and complexity of the motherboard design, as no additional memory chips are required on the board. There is usually some mechanism (via the BIOS or a jumper setting) to select the amount of system memory to use for graphics, which means that the graphics system can be tailored to only use as much RAM as is actually required, leaving the rest free for applications. A side-effect of this is that when some RAM is allocated for graphics, it becomes effectively unavailable for anything else, so an example computer with 512 MB RAM set up with 64MB graphics RAM will appear to the operating system and user to only have 448 MB RAM installed.

The disadvantage of this design is lower performance because system RAM usually runs slower than dedicated graphics RAM, and there is more contention as the memory bus has to be shared with the rest of the system. It may also cause performance issues with the rest of the system if it is not designed with the fact in mind that some RAM will be 'taken away' by graphics.

Other approach that gave similar final state is boost up of graphics, that is used in some SGi computers, most notably O2/O2+. The memory in these machines is simply one fast pool (2,1 GB/s per second in 1996) shared between system and graphics is done on demand including pointer redirection communication between main system and graphics subsystem. This is called Unified Memory Architecture


The first PC to use the SMA was the IBM PCjr, released in 1984. Video memory was shared with the first 128K of RAM. The exact size of the video memory could be reconfigured by software to meet the needs of the current program.

See also

* IBM PCjr
* Video memory
* Shared memory, in general, other than graphics

External links

* [,2542,t=SMA&i=51487,00.asp PC Magazine Definition for SMA]
* [ IBM PCjr information]

Wikimedia Foundation. 2010.

Look at other dictionaries:

  • Shared Nothing Architecture — Die Shared Nothing Architektur (SN) beschreibt eine Distributed Computing Architektur, bei der jeder Knoten unabhängig und eigenständig seine Aufgaben mit seinem eigenen Prozessor und den zugeordneten Speicherkomponenten wie Festplatte und… …   Deutsch Wikipedia

  • Memory architecture — describes the methods used to implement electronic computer data storage in a manner that is a combination of the fastest, most reliable, most durable, and least expensive way to store and retrieve information. Depending on the specific… …   Wikipedia

  • Shared memory — In computing, shared memory is a memory that may be simultaneously accessed by multiple programs with an intent to provide communication among them or avoid redundant copies. Depending on context, programs may run on a single processor or on… …   Wikipedia

  • Shared memory — (dt. „geteilter Speicher“) bezeichnet eine bestimmte Art der Interprozesskommunikation (IPC). Bei dieser Art nutzen zwei oder mehrere Prozesse einen bestimmten Teil des Hintergrundspeichers (RAM) gemeinsam. Für alle beteiligten Prozesse liegt… …   Deutsch Wikipedia

  • Unified Memory Architecture — Als Unified Memory Architecture (UMA) bezeichnet man Rechnerarchitekturen, in denen verschiedene Komponenten (z. B. die CPU und der Grafikprozessor) zusammen auf einem gemeinsamen Speicher arbeiten, anstatt dedizierten Speicher zu besitzen.… …   Deutsch Wikipedia

  • Distributed shared memory — (DSM), in Computer Architecture is a form of memory architecture where the (physically separate) memories can be addressed as one (logically shared) address space. Here, the term shared does not mean that there is a single centralized memory but… …   Wikipedia

  • Memory coherence — is an issue that affects the design of computer systems in which two or more processors or cores share a common area of memory.[1][2][3][4] In a uniprocessor system (whereby, in today s terms, there exists only one core), there is only one… …   Wikipedia

  • Memory ordering — is a group of properties of the modern microprocessors, characterising their possibilities in memory operations reordering. It is a type of out of order execution. Memory reordering can be used to fully utilize different cache and memory banks.… …   Wikipedia

  • Memory management — is the act of managing computer memory. The essential requirement of memory management is to provide ways to dynamically allocate portions of memory to programs at their request, and freeing it for reuse when no longer needed. This is critical to …   Wikipedia

  • Memory-mapped file — A memory mapped file is a segment of virtual memory which has been assigned a direct byte for byte correlation with some portion of a file or file like resource. This resource is typically a file that is physically present on disk, but can also… …   Wikipedia