control section to control and coordinate all processing activity, it must be able to locate each instruction and data item in storage. About now, you are probably wondering how the control section is able to find these instructions and data items. To understand this, let's look at storage as nothing more than a collection of mailboxes.">
TYPES OF INTERNAL STORAGE
You already know that the internal storage section is the holding area in which instructions and data are kept. For the control section to control and coordinate all processing activity, it must be able to locate each instruction and data item in storage. About now, you are probably wondering how the control section is able to find these instructions and data items. To understand this, let's look at storage as nothing more than a collection of mailboxes. Each mailbox has a unique address and represents a location in memory as shown in figure 2-2. Like the mail in your mailbox, the contents of a storage location can change, but the number on your mailbox or memory address always remains the same. In this manner, a particular program instruction or data item that is held in storage can be located by knowing its address. Some computers can address each character of data in memory directly. Others address computer words which contain a group of characters at a single address. Each computer word contains a group of characters at a single address. Some of the more common types of internal storage media used in today's computers are as follows: magnetic core, semiconductor, and bubble.
Figure 2-2. - Memory locations.
MAGNETIC CORE STORAGE
Although magnetic core storage is no longer as popular as it once was, we will cover it in some detail because its concepts are easily understood and apply generally to the more integrated semiconductor and bubble-type memories. Magnetic core storage is made up of tiny doughnut-shaped rings made of ferrite (iron), that are strung on a grid of very thin wires (fig. 2-3). Since data in computers is stored in binary form (refer to NEETS, module 13), a two-state device is needed to represent the two binary digits (bits), 0 for off and 1 for on. In core storage, each ferrite ring can represent a 0 or 1 bit, depending on its magnetic state. If magnetized in one direction, it represents a 1 bit, and if magnetized in the opposite direction, it represents a 0 bit. These cores are magnetized by sending an electric current through the wires on which the core is strung. It is this direction of current that determines the state of each core.
Figure 2-3. - Two-state principle of magnetic storage.
SEMICONDUCTOR STORAGE (THE SILICON CHIP)
Semiconductor memory consists of hundreds of thousands of tiny electronic circuits etched on a silicon chip (fig. 2-4). Each of these electronic circuits is called a bit cell and can be in either an off or on state to represent a 0 or 1 bit, depending on whether or not current is flowing in that cell. Another name you will hear used for semiconductor memory chips is integrated circuits (ICs). Developments in technology have led to large scale integration (LSI), which means that more and more circuits can be squeezed onto the same silicon chip. Companies are even manufacturing very large scale integrated circuits (VLSI), which means even further miniaturization.
Figure 2-4. - A semiconductor memory chip (integrated circuit).
Some of the advantages of semiconductor storage are fast internal processing speeds, high reliability, low power consumption, high density (many circuits), and low cost. However, there is a drawback to this type of storage. It is volatile, which means all data in memory is lost when the power supply is removed. Should the power on your computer fail and you have no backup power supply, all the stored data is lost. This is not the case with magnetic core storage. Core storage is nonvolatile. This means the data is retained even if there is a power failure or breakdown, since the cores store data in the form of magnetic charges rather than electric current.
One of the latest technological developments in storage media is the introduction of bubble memory. Bubble memory consists of a very thin crystal made of semiconductor material. The molecules of this special crystal act as tiny magnets (fig. 2-5). The polarity of these molecules or "magnetic domains" can be switched in an opposite direction by passing a current through a control circuit imprinted on top of the crystal. In this manner, data can be stored by changing the polarity of the magnetic domains. Since the principle is the same as for magnetic core storage, bubble memory is considered nonvolatile. The data is retained even if there is a power failure. Furthermore, the process of reading from bubble memory is nondestructive, meaning that the data is still present after being read. This is not the case with core storage, which must be regenerated after being read. If we were to view these magnetic domains under a microscope, they would look like tiny bubbles; hence the name, bubble memory.
Figure 2-5. - Bubble memory.
Q.9 Magnetic core storage is made up of what?