When the processor receives a request that it cannot fulfill from the cache, it sends a write buffer to return the unsatisfied block to main memory. So, what ought to be done here?

Hi, professor! I'm not sure about the following question. Consider a cache memory that provides three hundred 16-byte blocks. Now consider that you are processing all the data in a two-dimensional array of, say, four hundred rows by four hundred columns, using a pair of nested loops. Assume that the program stores the array column by column. You can write your program to nest the loops in either direction, that is, process row by row or column by column. Explain which way you would choose to process the data. What is the advantage? Conversely, what is the disadvantage of processing the data the other way? What effect does choosing the incorrect way have on system performance?

If the data being requested is too large to fit in the cache, a write buffer will bring it to the attention of the CPU by retrieving it from main memory. What should the next step be?

Think about the difference that exists between a cache that is totally associative and one that is directly mapped. This is a distinction that has to be made.

What should happen when a processor issues a request that cannot be satisfied by the cache while a block is being flushed back to main memory from the write buffer?

What is the effect on the cost of merging runs if the number of buffer blocks per run is increased, while keeping overall memory available for buffering runs fixed?

What is the point of having cache memory if we already have RAM, which is a memory that is built on volatile transistors? Is it feasible for a computer to use just a single kind of memory at a time?

What is the purpose of having cache memory if RAM (Random Access Memory) is based on transistors? Is it even possible to run every program on the same kind of computer memory?

The write buffer will return a block to main memory while the CPU submits a cache-unmet request. What may happen next?

cache.xls (or it can be a google sheet that you share with me in a cache.txt with the url in it )= a spreadsheet showing the cycle counts of a cached and non-cached machine. Also, you need to show the cache as the memory accesses progress in your program marking the hits and misses (misses in RED). You should have cells to show how many hits and how many misses your program has, and details on what is the cache replacement policy you implemented.   Here is code:   void c_cache_for_badge() {    int na = 8, nb = -3, nc = 10;    // pointers    int *npa;      npa = (int*)malloc(sizeof(int)*84);      // init integer pointers    for(int i = 0; i < 84; i++)    {        npa[i] = i;    }        // integer pointers    for(int i = 9; i < 70; i= i + 3)    {        nb += npa[i % 5] + npa[i % 3] + npa[i % 6];    }     }   void main() {    c_cache_for_badge(); }