Memory Organization | Computer Architecture

Memory Organization in Computer Architecture – The memory module of a digital system consists of two sections: an internal memory section and an external memory section. The internal memory is also called primary memory, as it can be directly accessed with random access.

Due to this property, such memories are able to respond fast enough to synchronize with the execution speed of the digital system processor. On the other hand, external memory devices can not be directly accessed by the processor. These devices are called secondary memory or auxiliary memory. 

Memory Hierarchy Organization

The total memory capacity of a digital system, like a computer, can be seen as a hierarchy of components. The computer system consists of all types of storage devices, from slow-bus high-capacity auxiliary memory to a relatively faster main memory. Since there are several types of memories, a memory hierarchy system is developed. At the bottom of the hierarchy, there are the relatively slow magnetic tapes used to store removable files.

Next are the magnetic disks used as backup storage. The main memory occupies a central position by being able to communicate directly with the CPU and with secondary memory devices through an I/O processor. A very high-speed memory is introduced between the main memory and the CPU to make the system fast. This is called cache memory. This makes the speed 5 to 10 times faster. The cache memory is very small, relatively expensive, and has very high access speed. 

The overall goal of using a memory hierarchy is to obtain the highest possible access speod with minimizing the total cost of the entire memory system. The memory hierarchy system is shown figure below.

Processor V/s Memory Speed

Auxiliary and cache memories are used by the computer system in addition to main memory. The cache memory stores the parts of the program and data that are most heavily used, while auxiliary memory stores those parts that are presently used by the CPU. The CPU can access cache and main memory directly, but not the auxiliary memory.

Therefore, the data from auxiliary memory is transferred to the main memory in large blocks of data by means of direct memory access (DMA). This makes the computer system slow. The cache memory is the fastest memory among all types of memories. The typical access time ratio between cache and main memory is about 1 to 7. Auxiliary memory average access time is usually 1000 times that of main memory. 

Many operating systems permit the CPU to process a number of independent programs simultaneously. This is called multiprogramming. The concept of multiprogramming provides the existence of two or more programs in different parts of the memory hierarchy at the same time. In this way, it is possible to keep all parts of the computer busy by working with several programs in sequence.

Indirectly, this concept of multiprogramming increases the execution speed of the CPU. Since more than one programs are executed simultaneously, all programs can not reside in main memory. A program with its data normally resides in auxiliary memory. When it is to be executed, it is transferred to the main memory. The memory management system supervises the flow of information between main memory and auxiliary memory.