Computer Architecture

profilejohn_usa12

Memory Access Impact on Performance:                                                                  
Assume that main memory accesses take 80 ns and that memory accesses are 42% of all instructions.  The L1 cache has a miss rate of 9% and a hit time of 0.58 ns.

1.Assume that the L1 hit time determines the cycle time for the processor.  What is the clock rate?

 

1._______

2.What is the Average Memory Access Time for the processor?

 

2.________

3.Assuming a base CPI of 1.0 without any memory stalls (once the pipeline is loaded), what is the total average CPI for the processor?

 

3. _______

 

We will consider the addition of an L2 cache to try to reduce the average CPI; on a miss, P1 will now first check L2 cache, and only if that is a miss, will then need a main memory access. The L2 miss rate is 85%, and L2 hit time is 4.88ns

 

4.What is the AMAT for the processor, with the inclusion of the L2 cache?


4._________

5. Assuming a base CPI of 1.0 without any memory stalls,  and using the same instruction miss as part 1 of this question, what is the total CPI (for all instruction types) for P1 with the addition of an L2 cache?

 

 

5._________

 


Question #1: The cache for this problem has 16 words.  You are going to evaluate the cache performance based on different mapping schemes, and different size blocks, to try to come up with the best mapping and block arrangement for this series of memory access calls.

 

Your mappings will be:
16 one-word blocks, Direct Mapping
16 one-word blocks, Fully Associative Mapping
4 4-word blocks, Direct Mapping
4 4-word blocks, Fully Associative Mapping
4-way Set-Associative, 16 one-word blocks

 

For each of the schemes,
1. Fill out the “top” table of “tags”
2. Show the placement of the block in the “bottom” table
3. Count the “hits” and “misses”
4. Compare the hit/miss ratios for the different mappings and block arrangements

 

If you were designing the cache, what do you think would have to most impact on the performance: larger block sizes, or different mapping schemes?  Explain your reasoning.




 

 

Word Address

Word

Bit Address

Direct 16
Tag

Direct 4
Tag

Fully Assoc.

16  Tag

Fully Assoc.

4

2

00000010

 

 

 

 

3

00000011

 

 

 

 

11

00001011

 

 

 

 

16

00010000

 

 

 

 

21

00010101

 

 

 

 

13

00001101

 

 

 

 

64

01000000

 

 

 

 

48

00110000

 

 

 

 

19

00010011

 

 

 

 

11

00001011

 

 

 

 

3

00000011

 

 

 

 

22

00010110

 

 

 

 

4

00000100

 

 

 

 

27

00011011

 

 

 

 

6

00000110

 

 

 

 

11

00001011

 

 

 

 


# hits : # misses                         _________     ________       _________       _________

CAT size in bits:

 

16 Blocks

Direct Map
1word blocks

Associative
1word blocks

4 Blocks

Direct Map
4-word block

Associative
4-word block

0  (0000)

 

 

 

 

0  (00)

 

 

1   (0001)

 

 

2  (0010)

 

 

3  (0011)

 

 

4  (0100)

 

 

 

 

1  (01)

 

 

5  (0101)

 

 

6  (0110)

 

 

7  (0111)

 

 

8  (1000)

 

 

 

 

2  (10)

 

 

9  (1001)

 

 

10 (1010)

 

 

11 (1011)

 

 

12 (1100)

 

 

 

 

3  (11)

 

 

13 (1101)

 

 

14 (1110)

 

 

15 (1111)

 

 

 


4-way Set Associative with 1-word blocks

Word Address

Word

Bit Address

4 way set
Tag

2

00000010

 

3

00000011

 

11

00001011

 

16

00010000

 

21

00010101

 

13

00001101

 

64

01000000

 

48

00110000

 

19

00010011

 

11

00001011

 

3

00000011

 

22

00010110

 

4

00000100

 

27

00011011

 

6

00000110

 

11

00001011

 


# hits : # misses                          ____________

CAT size in bits:                        ____________

 

4 way sets

Slot 1

Slot 2

Slot 3

Slot 4

 

 

0  (00)

 

 

 

 

 

 

1  (01)

 

 

 

 

 

 

2  (10)

 

 

 

 

 

 

3  (11)

 

 

 

 

 

 

 

  • 10 years ago
  • 20
Answer(1)

Purchase the answer to view it

blurred-text
NOT RATED
  • attachment
    cache_assignment_1.docx
Bids(0)