20.3. Semiconductor Memory

So, basically if you look memories are divided into mainly two types internal memory and external memory. So, internal memory basically is the semiconductor kind of a memory in which case you have a register. So, register is a part of the CPU itself. So, as we discussed in the last units, so this something like if you want to add two numbers so basically they have stored in a memory in a memory which is called the register; that means, something called a cache memory and a main memory. So, actually main memory is the word we have always heard the word called RAM. So, in a lay man language RAM, there are lot of technicalities we will come into, but in a lay man language what is known as a RAM is basically your main memory.

Memory in computers is primarily categorized into two types: internal and external memory. Internal memory refers to semiconductor memory used for quick access and processing by the CPU itself. This includes components like registers, cache memory, and main memory (often referred to as RAM). Registers are small storage locations within the CPU that temporarily hold data for processing, while RAM is where active programs and data are stored for quick access. In contrast, external memory refers to storage options like hard disks that are slower to access and used for long-term data storage.

Think of internal memory as your desk at home where you keep all your frequently used items like your laptop, notebooks, and important papers handy for immediate use. This is analogous to the CPU's registers and RAM. On the other hand, external memory is like your attic or storage room where you keep less frequently used things. You can retrieve items from there, but it takes more time to access them.

So, basically your CPU or your arithmetic logic unit of the main which is the computing unit of the CPU, basically it can talk only to the main memory that is it can generate the address and then it can read and write data from the main memory.

The main function of the CPU is to process data, which is stored in the main memory (RAM). The CPU generates addresses that point to specific data locations in the RAM, allowing it to read data for processing or write back results. This interaction forms the backbone of program execution in a computer, as the CPU needs to constantly interact with the main memory to access the code and data it will operate on.

Imagine you are a chef in a kitchen (the CPU) and the pantry (main memory) is where all your ingredients (data) are stored. Before cooking, you need to get ingredients from the pantry based on the recipe (addresses) you have. You cook using those ingredients, and sometimes, you may need to put back what you've made into the pantry. A good chef knows exactly where to find everything in the pantry, just like a CPU knows where to read and write data in RAM.

But there is another memory which lies in between the CPU and the main memory is called the cache memory. So, we will learn in more details about cache memory when we will going to into the full module on memory design. But the basic idea is that whenever you want to refer to some data, generally the address is generated for the main memory and as main memory is much slower compared to a register there is something in between which is the cache.

Cache memory is a smaller and faster type of volatile computer memory that provides high-speed data access to the CPU. It acts as a buffer between the CPU and the main memory to speed up data retrieval. When the CPU needs data, it first checks the cache memory. If the data is not found there (a cache miss), it fetches it from the main memory, which is slower. This makes accessing frequently used data quicker and enhances overall performance.

Consider cache memory as a spice rack in the kitchen where you keep your most-used spices readily available. Instead of going to the pantry (main memory) every time you need a spice, you grab it quickly from the spice rack. This speeds up your cooking process, just as cache memory speeds up data access for the CPU.

So, now let us go into what will be next in the unit summary, which will be more important to us are basically the main memory. So, main memory is a semiconductor memory as I told you that there are two types basically one is RAM and one is ROM. So, the RAM is the random access memory, basically it is volatile; and ROM is the read only memory, but both RAM and ROM are basically random access only that is that is not a sequential access.

Main memory consists of RAM (Random Access Memory), which is volatile (loses data when power is off) and ROM (Read Only Memory), which is non-volatile (retains data without power). Both types can be accessed randomly, meaning any data can be retrieved without having to read through others sequentially. This allows for efficient data management and access speed, critical for program execution and overall computer performance.

Think of RAM like a chalkboard used for solving math problems—once you're finished, the chalkboard is wiped clean, requiring you to write everything again when it's next needed. In contrast, ROM is like a printed textbook that always holds the original information, regardless of whether it's being used or not.

So, generally CPU will generate an n bit address, because the memory will have some locations which can be done by the address. So, generally there are 2𝑛 locations, where n is the number of address lines.

The CPU communicates with memory by generating addresses that correspond to specific locations in RAM. These addresses are created in 'n' bits, allowing for '2^n' individual memory locations. For example, if a system has 16 address lines (n=16), it can address 2^16 (65,536) distinct memory locations. This enables the CPU to locate and access the exact data it needs efficiently.

Imagine each address as a house number on a street. The number of house numbers (addresses) based on how many bits are used defines how many houses (memory locations) you can have on that street. A street with 16 house numbers (n=16) has 65,536 unique homes!

And again, but generally it is very cumbersome to write these values. So we write in a hexadecimal notion that is 0000 hex that is the first memory location and FFFF is the last memory location.

To manage the complexities of memory addressing, hexadecimal notation is commonly used because it is more compact than binary. Instead of writing long binary strings, programmers can represent memory addresses from 0x0000 (the first location) to 0xFFFF (the last location). This allows for easier reading and manipulation of addresses during programming and troubleshooting.

Using hexadecimal notation is akin to using shorthand notes instead of writing full sentences when jotting down messages. It makes communication faster and easier, just as hexadecimal offers a more efficient way to work with memory addresses compared to lengthy binary.

--