Go to the Cache Simulator at the University of Massachusetts at : http://www.ecs.umass.edu/ece/koren/architecture/Cache/frame1.htm

Build a 4-way set-associative cache with a total of 16 entries. Set Repeat to 10 cycles, and disable Limit Query. The rest should be set by default.

Next you need to enter your own sequence of 40 random numbers. For that, use this URL to generate a sequence of 40 random 2-digit numbers: https://www.random.org/strings/?num=40&len=2&digits=on&unique=off&format=plain&rnd=new

Copy and paste the random numbers in to the Query Sequence box for Task A.

We are going to run 5 rounds of simulations, changing the # of sets from 1 to 2, 4, 8, and 16. The order doesnt matter, just change the set number and hit Show Cache. For each round, record the final miss and hit rates, and report your results below:

# Set	Miss Rate	Hit Rate
1
2
4
8
16

Which set(s) represent a direct-mapped cache?_____________ Which represent an n-way set associative cache?___________ (note: our book calls this set-associative cache) Which represent fully-associative cache?_________________ (note: our book calls this associative cache) Which has the best performance? DM, NWSA, or FA? (circle one) If one of these is clearly better, then why do we have other designs? Why not always use that one? Is there a disadvantage? Do not guess, find the answer in the book and use their wording.