Home > Articles


  • Print
  • + Share This
This chapter is from the book

Terms you'll need to understand:

  • Memory controller

  • Read-only memory (ROM) and erasable programmable ROM (EPROM)

  • Capacitor

  • Bus



Concepts you'll need to master:

  • Volatile versus nonvolatile memory

  • Megahertz (MHz) and nanoseconds (ns.)

  • Memory address

  • Synchronization to clock cycle (clock tick)

  • Memory module versus memory chip

  • Odd and even parity checking

Computers use many different microchips and processors, with perhaps the most familiar being the CPU and the main memory—what people commonly refer to as the computer's RAM. Memory is just a temporary place to store information until the CPU can get to it. This information can be program instructions, data, or both. A typical instruction might be a request to store a number or an event somewhere. Another might be to retrieve that information from a particular place—an address. Volatile memory can only hold information when a normal electrical current is present. Nonvolatile memory can hold information without any electrical current.


Volatile, from the Latin "to fly," means that information "flies away" when there's no electricity to keep it in place. Television reporters often refer to an explosive situation as a volatile situation, meaning that it could change at any second. Nonvolatile memory, because it is not volatile, stays the same without any need for electricity.

Conceptual Overview

Memory is fairly easy to understand, once you've grasped the basic concepts. In a nutshell, a CPU moves bits of data into registers (storage places inside a chip). After it's dealt with these data bits to its satisfaction, the CPU works together with a memory controller to move the results out to memory cells (storage places on a memory chip). Both registers and memory cells have memory addresses, and every time a bit of data goes somewhere, it crosses a bus of some kind. That's it! Now go pass the exam.

All right, so it's a bit, so to speak, more complicated than that. Most memory began as dynamic random access memory (DRAM), and the main changes have been to either speed up the memory to match the CPU, or to speed up the CPU to match the memory. For the most part, the history of memory development revolves around synchronizing these two subsystems.


When we refer to speeding up memory, this usually means increasing either the actual speed of the chips or increasing the clock speed of associated buses.

Memory involves several basic concepts, the first of which is a grid or matrix. Because of this, we're going to put Table 3.1 to a slightly different use, making it a sort of "mind map." If you can see the way the overall concepts break down on a grid, then perhaps they'll be easier to remember.


A matrix is nothing more than an arrangement of columns and rows, like a spreadsheet or an Etch-a-Sketch. Columns go across the page, and rows go down the side. Cells going left to right (horizontally) have an X coordinate. Cells going up and down (vertically) have a Y coordinate. The direction of rows or columns is called the axis. Combining both the X and Y coordinates gives us an address in the grid, like a cell address in a spreadsheet.

Table 3.1 Mind map of basic memory concepts.

Data Storage – Nonvolatile

Disks Temporary swap files

ROM – Nonvolatile BIOS

Programmable ROM CMOS – volatile (trickle charge)

Flash BIOS – Nonvolatile

RAM – Volatile

Types of RAM

Types of Packaging

Main Memory













Cache Memory

L-1 and L-2

Card Modules

  • + Share This
  • 🔖 Save To Your Account