Log In
0 votes

if there is a miss in cache and hit in mm what is access time 60ns or 60+20=80ns.

how is the statement " reference started again" related to the  meaning of the question?

in CO and Architecture 122 views
"Reference started again" simply means that the data will be again accessed next time and thus, for the Miss penalty time, we need to take into account only till the data is moved into the cache from main memory.

1 Answer

0 votes

Since nothing is mentioned, I would assume Hierarchical access.

Tavg = H1T1 + (1-H1)H2(T1+T2) + (1-H1)(1-H2)(T1+T2+T3)

Since there are 3 levels. 

This formula can be interpreted as

[(Hit in Level 1) * (Time to access in level 1)] + [(Miss in Level 1) *  (Hit in Level 2) * (Time to access Level 1 memory + Time to access Level 2 memory)] + [(Miss in Level 1) * (Miss in Level 2) * (Time to access Level 1 memory + Time to access Level 2 memory + Time to access Level 3 memory)]

In the last term, H3 is assumed to be 1 as the word has to be there in the last level memory.

Thus in this case,

H1 = 0.9

H2 = 0.6

T1 = 20

T2 = 60

T3 = 12 * 106 ns

So, Tavg = 480,026 ns

edited by

Related questions

0 votes
1 answer
Consider the following statements: (i) Accessing of data in a column wise fashion maintains spatial locality only when the block size is equal to the total size of the elements in the row (ii) Coherence in write through protocol never occurs even cache memory is organized in multilevel. Which of the above is true?
asked Jan 9, 2019 in CO and Architecture Jay Bhutada 1 107 views
2 votes
0 answers
2 votes
1 answer
Suppose that in 250 memory references there are 30 misses in first level cache and 10 misses in second level cache. Assume that miss penalty from the L2 cache memory are 50 cycles. The hit time of L2 cache is 10 cycles. The hit time of the L1 cache is 5 cycles. If there are 1.25 memory references per instruction, then the average stall cycles per instruction is ________.
asked Jan 24, 2017 in CO and Architecture Pankaj Joshi 241 views