I do not want the whole lab. I need the results and how they came to be. Also how the parameters varied from 1 - 6, and a graph plotting the results.
This lab is on cache memories and you will need to work on timberlake to complete this project. The objective is to evaluate different cache organizations using the cache simulator called dinero. You will need a memory trace to complete this assignment. The memory trace information is provided at the end. You can use dinero on timberlake. via the command "dineroIV filename.din'' 1. Set dinero as a unified cache, and use block size 32, associativity 1 and write-back options. Vary the cache size as lk, 2k, 4k, 8k, 16k and 32k and plot the number of misses for each of the two traces (mm.32, mm.64). The trace files can be found in the subdirectory testing 2. Set instruction cache size 8192 and data cache size 8192 and vary the block size as 8, 16, 32 and 64 using the write back protocol. Generate a plot showing hit ratio versus block size. 3. Set instruction cache size 8192 and data cache size 8192 and vary the associativity as 1 2, 4 and full using the write back protocol. Generate a plot showing the hit ratio versus the associativity 4. Same as 3 but use write-through. 5. Set instruction cache size 8192, data cache size 8192, block size 32 and associativity level 4 and vary the replacement policy using LRU, FIFO and random. Generate a plot showing the hit ratio versus the replacement policy. 6. Same as 5 but use full associativity This lab is on cache memories and you will need to work on timberlake to complete this project. The objective is to evaluate different cache organizations using the cache simulator called dinero. You will need a memory trace to complete this assignment. The memory trace information is provided at the end. You can use dinero on timberlake. via the command "dineroIV filename.din'' 1. Set dinero as a unified cache, and use block size 32, associativity 1 and write-back options. Vary the cache size as lk, 2k, 4k, 8k, 16k and 32k and plot the number of misses for each of the two traces (mm.32, mm.64). The trace files can be found in the subdirectory testing 2. Set instruction cache size 8192 and data cache size 8192 and vary the block size as 8, 16, 32 and 64 using the write back protocol. Generate a plot showing hit ratio versus block size. 3. Set instruction cache size 8192 and data cache size 8192 and vary the associativity as 1 2, 4 and full using the write back protocol. Generate a plot showing the hit ratio versus the associativity 4. Same as 3 but use write-through. 5. Set instruction cache size 8192, data cache size 8192, block size 32 and associativity level 4 and vary the replacement policy using LRU, FIFO and random. Generate a plot showing the hit ratio versus the replacement policy. 6. Same as 5 but use full associativity
