4.3.5. Magnetic Disks

Magnetic disks or hard disks are far cheaper; about 1000 times cheaper than DRAMs being only about 0.2 to 2 dollars per GB. However, it is also about 1000 times 100 times, 100 to 1000 times slower than DRAM units. Its access times ranges in between 5 to 20 milliseconds.

Magnetic disks, also known as hard disks, are storage devices used in computers to save data. They are significantly more affordable than DRAM (Dynamic Random Access Memory), costing only between $0.2 to$2 per gigabyte, making them an economical choice for large data storage. However, they are much slower, with access times between 5 to 20 milliseconds, compared to the speed of DRAM, which can access data in nanoseconds. This means retrieving data from a magnetic disk can take thousands of times longer than from DRAM.

Think of it like a library: magnetic disks are the vast archives containing all the books (data) where you can find any book you want for a low fee. However, retrieving a specific book takes a lot longer because you need to walk through the aisles (slower access times) compared to grabbing a book that's right next to you in your backpack (like DRAM).

So, to achieve the best performance what would we desire? We would desire a very large capacity memory which can hold all our programs and data and which works at the pace of the processor. However, in practice we saw the cost and performance parameters and it is difficult to achieve.

To enhance computer performance, it's ideal to have a memory system that is both large enough to store all the necessary programs and data, and fast enough to keep up with the processor's speed. However, achieving this balance is a challenge because as memory capacity increases, typically so do costs, and faster memory types like SRAM are significantly more expensive compared to slower memory types like magnetic disks.

Imagine if you want to run a restaurant smoothly – you need a huge kitchen (large memory) that can prepare dishes quickly (fast access), but having a gigantic kitchen with state-of-the-art equipment (high cost) is not always feasible due to budget constraints.

So, although the faster memories, the mem although SRAMs are very fast in terms of access time, they are also very costly. The solution is to have memory hierarchy where smaller more expensive and faster memories are supplemented by larger, cheaper and slower memories.

To optimize both performance and cost, systems are organized into a memory hierarchy. This hierarchy comprises layers of memory, where fast but expensive types like SRAM are used in smaller quantities at the top, and larger, slower types like magnetic disks are used at the bottom. This setup helps to speed up data access while keeping costs manageable. As you go lower in the hierarchy, the memory becomes cheaper but slower.

Think of packing for a trip: you can take a small, high-quality suitcase (expensive and fast), but for bulk items, you might need larger, cheaper bags to hold everything, even if it takes more effort to get to the items at the bottom.

Principle of the locality of reference is based on the fact that programs tend to access data and instructions in clusters, in the vicinity of a given memory location.

The principle of locality of reference suggests that programs typically access memory by utilizing small clusters of instructional or data blocks repeatedly. This principle supports the design of memory systems because if the data required again is likely close to the initial access point, then putting frequently accessed data in faster memory improves efficiency.

This is like finding your favorite snacks in a kitchen: if you always grab the chips from the same cabinet, it makes sense to keep that cabinet organized and close at hand rather than buried in a distant pantry.

So, principle of locality makes hierarchical organization of memory possible. So, how can we do that? For example, we can store everything we store everything in the magnetic disk. And then we copy recently accessed and nearby data in a small DRAM memory or the main memory.

Leveraging the principle of locality of reference allows computers to utilize their memory hierarchy effectively. By storing all data on a slower magnetic disk and maintaining copies of frequently accessed data in faster SRAM (cache) or DRAM (main memory), the system can quickly retrieve the necessary information when needed without constantly accessing the magnetic disk.

Consider a chef who keeps their most-used ingredients in a small countertop container (cache) instead of going to the pantry (magnetic disk) every time they need something. This keeps cooking efficient without sacrificing storage space.

--