Unit 1 part 2: We ended part 1 of unit 1 by saying that we have different memory technologies which vary in terms of their access times and cost per GB. For example, we said that SRAMs are very fast and its speed is about one 0.5 to 2.5 nanoseconds, its access time; that means, it is on an average about one-tenth as fast as the processor ok. However, the cost per GB of this type of memories is also very huge. The cost per GB is about 2000 dollars to 5000 dollars.

Then we have DRAMs which are about 150 to 100 times slower than SRAMs; that means, to bring a certain amount of data a data unit a word from DRAM the processor will require about hundreds of processor cycles to do so. The speed of a DRAM is typically in the range of 50 to 70 nanoseconds; that is the access time is in the range of 50 to 70 nanoseconds. But, it is also about hundred times cheaper than SDRAMs. So, the typical cost of DRAM units range in between 20 dollars to 75 dollars per GB.

Magnetic disks or hard disks are far more 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, 100-2000 times slower than DRAM units. Its access times ranges in between 5 to 20 milliseconds. So, to bring a data word from the hard disk, the processor required tens of thousands of processor cycles.

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. That means, if a processor requires a memory word in one cycle it is available in the processor from memory in the next cycle itself.

However, in practice we saw the cost and performance parameters and it is difficult to achieve. So, to achieve the greatest performance memory should be able to keep pace with the processor. It is not desirable to wait for instruction slash operands when the processor executes instructions. And hence we would like to use the fastest available memory technology. We also need a large capacity memory to hold all our required information. However, the cost of memory must be reasonable with respect to other components. Hence we understand that we have a design trade off. 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.

Principle of the locality of reference is based on the fact that programs tend to access data and instructions and data in clusters, in the vicinity in the near vicinity at a of a given memory location. So, programs access a small portion of memory at a given time. Why? Because programs typically contain a large number of loops and subroutines, and within a loop order subroutine a small set of instructions are repeatedly accessed. These instructions again tend to access data in clusters. So, there are two distinct principles in the locality of reference. Temporal locality which says that items access recently are likely to be accessed again. For example, the instructions within a loop. So, if the instructions within in one iteration of the loop will be again accessed in the next iteration of the loop. And special locality in items near those access recently are likely to be accessed soon; for example, sequential access of data from an array.

So, how does this locality how does this principle of the locality of reference helps to maintain this hierarchical memory organization. So, principle of locality makes hierarchical organization of memory possible. So, how can we do that? For example, we can store everything we stored everything in the magnetic disk. And then we copy recently accessed and nearby data in a small DRAM memory or the main memory. So, the main memory uses this technology of DRAM from the disk. So, we have the magnetic disk which stores everything and whatever we require currently we access, we access it and store it in a DRAM or the main memory. Then whatever is still more recently accessed data and instructions are stored in an SRAM memory which is cache from the DRAM.

Cache memory: So now we begin our discussion on cache memory. So, cache memory as we said is based on the SRAM memory technology. It is a small amount of fast memory which sits between the main memory and the CPU and it may be located within the CPU chip or a separate modules which are plugged in on the motherboard.