What is SDRAM?

internet computers

SDRAM is another of those powerful acronyms that describes a lot more than it sounds like it does. The letters stand for Synchronous Dynamic Random Access Memory, and it is a fast method of delivering computing capacity. SDRAM can run at 133 Mhz, which is much faster than earlier RAM technologies.

SDRAM is very protective of its data bits, storing them each in a separate capacitor. The benefit of this is the avoidance of corruption and the maintenance of "pristine" data. The drawback is that those same capacitors that are so useful at storing the SDRAM bits also happen to be very bad at keeping electrons in check; the result is where the Dynamic part of the name comes in, as refreshes are required to maintain data integrity. Once all of that dynamic refreshing and storing are done with, the result is a dense package of data, one of the densest in the business world.

We add the Synchronous part with a subroutine that lines itself up with the computer system bus and processor, so that all operations take place at the same time. Specifically, the computer's internal clock drives the entire mechanism. Once the clock sends out a signal saying that another unit of time has passed, the SDRAM chips go to work. In addition to the dense data package of DRAM, SDRAM allows a more complex memory pattern, giving you an extremely powerful method of storing and accessing data.

Another benefit of SDRAM is what is called pipelining. Because the SDRAM chips are so dense and complex, they can accept more than one write command at a time. This means that a chip can be processing one command while it accepts another one, even if that new command has to wait its turn in the pipeline. Previous RAM chips required proprietary access, allowing only one command at a time throughout the chip. In this way, SDRAM chips are faster than their predecessors.

This mostly describes single-data SDRAM chips, or SDR SDRAM. An even newer kind of chip is double-data-rate SDRAM, or DDR SDRAM. This allows for even greater bandwidth by making pipeline data transfers twice for every unit of time put forth by the computer's internal clock. One transfer takes place at the beginning of the new unit of time; the other takes place at the end.

SDRAM chips first came to the computing forefront in 1997. In just three years, they had become the dominant force in memory chips across the computing spectrum.

Related wiseGEEK articles

Category

New: Discuss this Article

Posted by: djfryoux
What does cache memory use? WRAM, SRAM, DRAM, or EDO?
Posted by: anon12701
Modern processor caches use SRAM.

FREE: Subscribe to wiseGEEK

 
    learn more

our strict privacy policy ensures that your email address will be safe



Written by David White

copyright © 2003 - 2008
conjecture corporation