Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

Q5 (3 marks) In this question, we investigate the effect of block size on averag

ID: 3602823 • Letter: Q

Question

Q5 (3 marks) In this question, we investigate the effect of block size on average memory access in a system with only one level of cache. The following table shows the miss rate versus block size of the cache. Assume that the memory system takes 80 clock cycles of overhead and then delivers 16 bytes every 4 clock cycles. Thus, it can supply 16 bytes in 84 clock cycles, 32 bytes in 88 clock cycles, and so on. Which block size has the smallest average memory access time assuming that hit time is 1 clock cycle (independent of block size)? why? Block size16 32 64128256 Miss rate 3.2 % | 3% | 2.6% | 2.3% | 1.9%

Explanation / Answer

Solution:

The formula is

H*(clock cycles) + (1-H)*(clock cyle)

for block size 16

0.968 * 1 + 0.032 * 84= 3.656 unit of time

for block size 32

0.97 * 1 + 0.03 * 88= 3.61

for block size 64

0.974 * 1 + 0.026 * 96= 3.47

for block size 128

0.977 * 1 + 0.023 * 112= 3.553

for block size 256

0.98.1 * 1 + 0.019 * 144=3.717

comparing all we can see that block size 64 byte is faster of them all.

I hope this helps, please let me know in case of any doubt. Thumbs up if this helped.