What Is Semiconductor Memory?

A memory cell is actually a type of sequential logic circuit. According to the type of memory used, it can be divided into read-only memory (ROM) and random access memory (RAM). The functions of the two are quite different, so they are also different in description ^[1] .

Memory is a memory component used to store programs and various data information. Memory can be divided into two categories: main memory (referred to as main memory or internal memory) and auxiliary memory (referred to as auxiliary or external memory). The main information is directly exchanged with the CPU. ^[2]

Computer memory can be divided into internal memory and external memory. The internal memory is frequently used by the computer during program execution and is directly accessible during one instruction cycle. External memory requires the computer to read information from an external storage device such as a magnetic tape or a magnetic disk. This is similar to students taking notes in class. If the student knows the content without looking at the notes, the information is stored in "internal memory". If the student must review the notes, the information is in "external memory". ^[3]

The storage medium constituting the memory mainly uses a semiconductor device and a magnetic material. The smallest storage unit in the memory is a storage cell of a bistable semiconductor circuit or a CMOS transistor or magnetic material, which can store a binary code. A storage unit is composed of several storage units, and then a memory is composed of many storage units. ^[4]

Memory structure In the MCS-51 series single-chip microcomputer, the program memory and data memory are independent of each other, and the physical structure is also different. Program memory is read-only memory and data memory is random access memory. From the physical address space, there are 4 storage address spaces, namely on-chip program memory, off-chip program memory, on-chip data memory and off-chip data memory. The I / O interface and the external data memory are uniformly addressed. ^[5]

In order to improve the performance of the memory, various memories with different storage capacities, access speeds, and prices are usually organized into a multi-layer memory according to a hierarchical structure, and the management software and auxiliary hardware are organically combined into a unified whole, so that the stored programs and data Distributed in each memory in a hierarchy. ^[6]

The storage capacity of any memory chip is limited. To form a certain amount of memory, a single chip often cannot meet the word length or the number of storage units, and even the word length and the number of storage units cannot meet the requirements. At this time, multiple memory chips need to be combined to meet the demand for storage capacity. This combination is called memory expansion. The problems to be solved during memory expansion mainly include bit expansion, word expansion, and word bit expansion. ^[7]

Flash memory controller functions include memory organization, boot selection, IAP, ISP, on-chip flash programming, and checksum calculation. Flash memory controller mapping and system memory mapping are introduced in the memory organization. The Flash memory controller contains on-chip Flash and Boot loader. The on-chip memory is programmable, including APRON, LDROM, data flash, and user configuration area. The address mapping includes Flash memory mapping and five address mappings: LDROM that supports IAP function, LDROM that does not support IAP function, APRON that supports IAP function, APROM that does not support IAP function, and a boot loader that supports IAP function. ^[8]

The purpose of storage management is to provide users with convenient, safe and sufficient storage space. ^[9]

How to increase the access speed of disks, how to prevent data loss due to disk failures, and how to effectively use disk space, has always been a problem for computer professionals and users; and large-capacity disks are very expensive, which is a big problem for users burden. The emergence of disk array technology solved these problems in one fell swoop. ^[10]

Memory is the main medium for data storage in the computer. With the development of recent years, changes in memory are changing with each passing day, and various new types of memory have entered the market, and the maintenance methods for new types of memory have become imminent. ^[11]

From PCRAM and MRAM to RRAM and more, a whole new series of memory technologies are constantly coming to fabs. What drives this process are technological advances in the gaming and mobile products arena, and the development of cloud computing. These applications are very important and they are continuously expanding the capabilities of today's mainstream storage technologies. For example, game applications require extremely fast main memory and high-capacity auxiliary (storage-type) memory to process data without the user being aware of it and quickly manage massive amounts of graphics data. After all, no one wants to suddenly encounter unexpected stutters when the game is at a critical moment. For cloud computing, its biggest advantage is the ability to access massive amounts of data through the network without having to store this data directly on our personal devices. Similarly, speed is critical, because nobody wants to wait for even one nanosecond unless necessary. ^[12]

With the rapid development of data storage technology, users have higher and higher requirements for storage cost performance, and cloud storage technology does not require the support of hardware equipment, which greatly increases the security performance of storage, and users do not need to maintain hardware facilities Reduced investment costs and improved storage efficiency. ^[13]

Semiconductor memory