Computer systems rely on two different types of storage, primary and secondary. Primary storage is fast and usually volatile, communicating directly with the CPU. Secondary storage has more capacity and is permanent but slower, usually communicating directly with primary storage.

Random access memory (RAM) is the primary storage method of most modern day computers. RAM is basically many, many electronic switches. This means that RAM is extremely fast. Most RAM is also volatile, "RAM chips lose their contents if the current is turned off or disrupted (as in a power surge, brownout, or electrical noise generated by lightning or nearby machines)" (Stair & Reynolds, 2006, 51).

Most personal computers use one of two types of RAM; (1) dynamic random access memory (DRAM), or (2) Synchronous dynamic random access memory (SDRAM). Older computers used DRAM whereas today's computers use some version of SDRAM. The main use for RAM is temporary storage buffer for the processor(s). Most data is stored in secondary storage which is slow. The CPU requests instructions (code) from RAM which in turn has to retrieve the data from secondary storage if the data is not already in the RAM's storage.

In general, more RAM means greater performance, but if a computers secondary storage devices are very slow, the RAM will have to wait for the device. This is due to the ability to store more information in faster memory. An example of this phenomenon occurs when loading a large spreadsheet or document. The RAM is waiting for the secondary storage device thus creating a lag or delay.

RAM should not be confused with another type of random access memory, non-volatile RAM (NVRAM). NVRAM has the advantage of keeping its contents when power is lost. NVRAM is generally slower than RAM, but recent technology advances show promise for both capacity and speed increases. The most common type of NVRAM in use today is a flash drive. These drives are very slow, but made for portability.

If computer systems were limited to primary storage, which early computers oftentimes were, data processing would be much more difficult. Large datasets would need to be pre-processed to a summary form and processed yet again to get the final data.

To get around this problem secondary or permanent storage devices were created. Historically these were punch cards and magnetic tape. Modern versions of secondary storage devices include; hard and floppy disks, CD ROM's and DVD's, and tape.

In the early 1980s the first form of permanent storage on many first generation personal computers was the floppy disk. The most common sizes were 8, 5.25 and 3.5 inches. Data capacity varied, but the standard for 5.25" disks was 180KB and was used by Apple for it's Apple 2 product. Tandon for use in IBM's (then) new PC and was 360KB.

Unless legacy data is stored on a floppy disk, no compelling use case exists for their use. The cost per gigabyte is very high, about $347 (Stair & Reynolds, 2006, 53), and floppy drives are extremely slow.

Hard disks are the major storage devices on computers today. Hard drives are extremely reliable, relatively fast and cheap. The cost per gigabyte is around $0.30. Modern hard drives have capacities in excess of 500 gigabyte and can be grouped together to form a redundant array of inexpensive disks (RAID). These RAID's can be configured to boost reliability, speed and capacity.

The only downside to hard disks is the lack of portability. Solutions exist such as hard drive caddies or USB connected drives, but outside the music and film world zip drives have become the first choice when data portability is needed.

CD ROM is an optical disk format with a capacity of 700 MB. The data can not be modified once written to the media. CD versions exist which allow for writing multiple times (CD R/W), but the disk capacity slowly diminishes with every write cycle. CD-ROM is a very good option for short term archiving of data, with an expected lifespan of "between two to five years, depending on the quality of the CD" (Blau, 2006).

CDs are rapidly being replaced by digital versatile disc (DVD). DVDs have much greater capacity, 4.7GB on a single side. The cost per gigabyte is around $0.29 for DVD and $0.34 for CD ROM. The one disadvantage of DVD technology is two competing standards, DVD-RW and DVD+RW. In practice, all modern players and recorders can use either system. Historically DVD-RW was considered the most compatible.

Magnetic tape has long been the media of choice for backing up systems and data. Just like cassette tapes, "magnetic tape is a Mylar film coated with iron oxide. Portions of the tape are magnetized to represent bits" (Stair & Reynolds, 2006, 53). Magnetic tape has a life span of greater than 30 years, but care must still be taken to avoid degradation and accidental erasure.

The principle drawback of tape is the slow speed at which it reads and writes. Magnetic tape is a sequential media as opposed to a random access media like hard and floppy disks and CDs. If a piece of data is needed in the middle of the tape, no way exists to jump to the needed section. The tape must either go forward or rewind to the proper point. This drawback limits the current use magnetic tape to archival purposes, although historically tape was the media of choice for data processing jobs.

Primary and secondary storage serve two distinct functions. Primary storage is the highway to the processor while secondary storage serves as a permanent storage location. When deciding on which technology to use understanding not only the terminology, but also the most appropriate use of the technology is required.

Blau, J. (2006, January 10). Do Burned CDs Have a Short Life Span?. Retrieved October 9, 2007, from http://msn.pcworld.com/article/id,124312-page,1/article.html

Stair, R. M., & Reynolds, G. W. (2006). Fundementals of Information Systems (3rd ed.). Boston: Thomson Course Technology.