RAM & Cache

Cache memory  is a high-speed, small-sized type of volatile computer memory that plays a crucial role in improving CPU (Central Processing Unit) performance. It serves as a bridge between the CPU and the slower, larger main memory (RAM). The CPU registers form part of the cache memory.

The primary purpose of cache memory is to reduce the time it takes for the CPU to access frequently used data and instructions, thereby enhancing overall system performance. 

Cache memory is static ram (SRAM) that stores its data in flop flop circuits.

Cache is usually split into levels: L1, L2, and L3, each with different sizes, speeds, and proximity to the CPU core.

• Closest to the CPU core (often built into it).

• Smallest (typically 16KB–128KB per core).

• Fastest (latency: ~1-2 CPU cycles).

• Often split into:

  • L1d (Data) cache

  • L1i (Instruction) cache

• Used for: Storing the most frequently used data and instructions.

• Because of its speed, it's accessed first by the CPU.

• Larger than L1 (usually 128KB–1MB per core).

• Slower than L1 but still very fast (latency: ~3–15 cycles).

• May be dedicated per core or shared between a few cores.

• Used for: Storing data that isn’t in L1 but is still likely to be reused soon.

• Largest (usually 2MB–64MB, shared among all CPU cores).

• Slowest of the three (latency: ~20–50+ cycles), but still faster than RAM.

• Shared across cores on the same processor (helps with inter-core communication).

• Used for: Reducing access times to DRAM and improving performance in multi-core tasks.

A cache hit occurs when the CPU requests a piece of data or instruction, and that data is found in the cache memory. In other words, the data needed by the CPU is already present in the cache.

A cache miss occurs when the CPU requests data or instructions that are not present in the cache memory. In this case, the cache does not contain the required data, and the CPU must fetch it from a slower memory hierarchy level, typically from the main memory (RAM).

Cache hits are desirable because they result in very fast memory access times. The CPU can retrieve the required data directly from the cache, avoiding the longer latency associated with accessing data from main memory (RAM) or even slower storage devices like hard drives or SSDs.

Advantages

• Cache memory is faster than RAM memory because it is located directly on the CPU using flip-flop circuits.

Disadvantages

• Cache memory is more expensive than RAM because the circuitry is more complex than RAM.
• It is also more energy intensive, due to the flip flop circuits being constantly active.

RAM (Random Access Memory) is a type of volatile computer memory that serves as the primary workspace for actively used data and instructions within a computer system. It utilises dynamic ram DRAM, a form of memory that stores data in capacitors. This form of memory is much cheaper than static ram, allowing much larger amounts of data to be stored.

RAM is much faster than other storage devices like hard drives and SSDs. It provides the CPU with rapid access to data and instructions, enabling quick data retrieval and manipulation. This speed is crucial for the efficient operation of applications and the overall responsiveness of the computer.

Without RAM the computer would have to load data in directly from the HDD, which has significantly slower speeds. This would lead to bottlenecks in processing that would slow down the operation of the computer to the point where it would be unusable.

During Bootup

When you power on your computer or restart it, the computer's BIOS or UEFI firmware initializes hardware components and loads the operating system (e.g., Windows, macOS, Linux) from non-volatile storage (e.g., a hard drive or SSD) into RAM. The operating system's kernel and essential system files are loaded into RAM during this process.

When software is opened

When you launch a software application (e.g., a web browser, word processor, or video game), the operating system loads the necessary program files and libraries from storage into RAM. This includes executable code, user interface elements, and data structures needed to run the application.

When Files are accessed

When you open a file (e.g., a document, image, or video), the relevant portions of that file are read from storage and loaded into RAM. 

In cases where physical RAM is limited, modern operating systems use a technique called virtual memory. They allocate a portion of the storage device (e.g., a hard drive or SSD) as virtual RAM.

When physical RAM becomes scarce, less frequently used data is moved to this virtual memory, allowing more critical data to remain in physical RAM.

This process, known as paging or swapping, enables efficient memory management.

Speed

RAM is significantly faster than secondary storage devices like hard drives and SSDs. It provides quick access to data and instructions, which is essential for the rapid execution of applications and tasks.

Data Access

RAM offers random access, which means that any part of the memory can be accessed quickly with nearly the same speed. This is important for efficient data retrieval and manipulation.

Cost

The DRAM used in RAM is cheaper than the SRAM used in cache memory and so can be installed in much greater quantities on a computer system.

Volatility

RAM is volatile memory, meaning that it loses its data when the computer is powered off or restarted. This makes it unsuitable for long-term data storage.

Limited Capacity

RAM has limited capacity compared to secondary storage devices. While you can have multiple gigabytes of RAM, it's still relatively smaller than the storage capacity of hard drives and SSDs, limiting the amount of data that can be held in RAM.

Cost

RAM can be expensive, especially as you aim for larger capacities and higher speeds. This can be a significant cost factor in building or upgrading a computer system.