Exercise 1.14 Section 1.8 cites as a pitfall the utilization of a subset of the performance equation as a performance metric. To illus trate this, consider the following data for the execution of given instruction sequence of 106 instructions in different processors.

Processor Clock rate CPI P1 4 GHz 1.25 P2 3 GHz 0.75 1.14.1 [5] <1.8> One usual fallacy is to consider the computer with the largest clock rate as having the large performance. Check if this is true for P1 and P2.

1.14.2 [10] <1.8> Another fallacy is to consider that the processor executing the largest number of instruction will need a larger CPU time. Considering that processor P1 is executing a sequence of 106 instructions and that the CPI of processors P1 and P2 do not change, determine the number of instructions that P2 can execute in the same time that P1 needs to execute 106 instructions.

1.14.3 [10] <1.8> A common fallacy is to use MIPS (millions of instructions per second) to compare the performance of two different processors, and consider that the processor with the largest MIPS has the largest performance. Check if this is true for P1 and P2.

Another common performance fi gure is MFLOPS (million of fl oating-point operations per second), defi ned as MFLOPS = No. FP operations/execution time × 106 but this fi gure has the same problems as MIPS. Consider the programs in the following table, running on a processor with clock rate = 3 GHz.

Instr. count L/S instr. FP instr. Branch Instr. CPI(L/S) CPI(FP) CPI(Branch)

a. 106 50% 40% 10% 0.75 1 1.5

b. 3 × 106 40% 40% 20% 1.25 0.70 1.25 1.14.4 [10] <1.8> Find the MFLOPS fi gures for the programs.
1.14.5 [10] <1.8> Find the MIPS fi gures for the programs.
1.14.6 [10] <1.8> Find the performance for the programs and compare with MIPS and MFLOPS.