|
Same modification as above. Modified formulas from 3: Read_time_for_this_cache will be R = RH+miss_rate*RLL Write_time_for_this_cache will be W = WH + (1 – %wb)*WLL 9. A cache is write-back, write-allocate, with write buffer to a lower level We write to lower level every time we evict a dirty block and we pay for this D. Now with write buffer we will pay D only when we cannot use write buffer. Modified formulas from 4: Read_time_for_this_cache will be R = RH+miss_rate*(RLL + E) = RH+miss_rate*(RLL + %dirty*(1 – %wb)*D) Write_time_for_this_cache will be W =WH+miss_rate*(RLL + E) = WH+miss_rate*(RLL + %dirty*(1 – %wb)*D) 10. A cache is write-back, no-write-allocate, with write buffer to a lower level We write to lower level every time we evict a dirty block and we pay for this D. Now with write buffer we will pay D only when we cannot use write buffer. We also write to lower level every time we have a write miss, and we pay for this WLL. Now with write buffer we will pay WLL only when we cannot use write buffer.
Modified formulas from 5: Read_time_for_this_cache will be R = RH+miss_rate*(RLL + E) = RH+miss_rate*(RLL + %dirty*(1 – %wb)*D) Write_time_for_this_cache will be W = WH+miss_rate*(1 – %wb)*WLL ...For more information, Please download the word document. Word file "We have a CPU, L1 cache, L2 cache and a memory. We will look at different exampl" Free Download Address Click here to download the word document:We have a CPU, L1 cache, L2 cache and a memory.doc Note |