Test Unit 1 Computer Architecture UCLM (Albacete)

There are no two processors... With the same organization and different architecture. With the same architecture and different organization. None of the above. With the same architecture and different technology.

A computer hard disk is replaced with a new disk that is 10 times faster. Obtain the utilization for this device for the computer to run at twice the speed. 55%. 60%. 35%. None of the above.

Assume a base computer, Mbase, with a clock frequency of 1 GHz and with the following CPI for each of the three supported instruction types: A-2, B-4, and C6. An improvement is made in the processor of that computer, resulting in an optimized computer, Mopt, with a clock frequency of 1.2 GHz and with the following CPI for each type of instruction: A-2, B-3, and C-5. Assume the following mix of instructions for a given test program: A-30%, B-50%, and C20%. How much faster is Mopt than Mbase for this test program?. 47%. 147%. None is correct. 65%.

We have a 1.8 GHz PowerPC capable of running 700 MIPS and a 1.6 GHz Pentium 4 capable of running 850 MIPS. We can state that…. Pentium 4 offers the best performance. PowerPC offers the best performance. We need more information to compare the performance of both processors. None of the above.

We run a program with 5.5x10^9 instructions on a 3 GHz computer with a CPI of 0.9 cycles. The execution time for this program is…. 1.65 sec. None of the above. 14.85 sec. 16.5 sec.

From a compiler C we have obtained an improved compiler M, which provides code for floating-point instructions that is k times faster than the original. If 90% of total time is devoted to floating-point instructions when using compiler C, what would be the maximum speedup that we could obtain with compiler M?. 10. 100. 0.9. None of the above.

VLIW processors are those processors that... Compiler packs multiple operations in a single instruction. None of the above. Do not require to replicate hardware components to execute multiple instructions in each pipeline stage. Use multiple dependent functional units in the registers.

VLIW processors are those that... The compiler packages multiple operations into a single instruction. All the above. Use multiple processors. It is not necessary to replicate components to execute multiple instructions at each stage.

Inés's computer costs 650€ and Irene’s computer costs 999€. Inés obtains the result of a program in 30 minutes, whereas Irene obtains the same result after 20 minutes. We define cost effectiveness as the relationship between the performance observed and the cost of the computer. Inés's computer is more cost effective in this case. Inés's computer will be always more cost effective. Irene's computer is more cost effective in this case. Irene's computer will be always more cost effective.

MFLOPS is a metric that: All the options are correct. It is used in the TOP500. Is affected by the combination of integer and floating-point operations in the program executed. Is affected by the combination of floating-point operations of different types and the duration of the program executed.

Which of the following system elements is not part of the concept of computer organization?. Input/Output. Control signals. Memory technology used. Interfaces between computer and peripherals.

Which of the following system elements does not take part of the concept of computer architecture?. Control signals. Instruction set. Data representation. Input/Output.

Computer performance: Is inversely proportional to the execution time. Has no connection with the execution time. None of the above. Is inversely proportional to the square of the execution time.

The memory hierarchy system aspect of a computer belongs to: The computer organization concept. The computer architecture concept. The computer technology concept. It belongs to all of them.

The memory hierarchy system aspect belongs to: The computer organization concept. The computer architecture concept. The computer technology concept. It belongs to all of them.

Let us suppose that a program takes 150 seconds to run of which 100 are used in addition operations. How much do we have to improve the speed of the addition for the program to run at twice the speed?. 4. 6. 2. It is not possible.

Let's assume that we have improved a computer, so that all the floating point instructions are now executed four times faster. We are looking for a test program to measure the performance of the new computer, which must obtain a speedup of 2. If the execution time of this test program with the old floating point hardware is 100 seconds, how many seconds should be devoted to the execution of floating point instructions?. 66.6 seconds. 33.3 seconds. 50 seconds. None of the above.

Assume a computer with a clock frequency of 500 MHz and with the following CPI values for each of the four types of instruction supported (A, B, C, and D): CPI_A=2, CPI_B=3, CPI_C=4, CPI_D=5. Assume the following instruction mix for a given test program: A: 40%, B: 25%, C: 25%, D: 10%. What would be the MIPS value for this computer?. 163.9 MIPS. 170.1 MIPS. 150.2 MIPS. None of the above.

Those system attributes that are visible to the programmer are part of the:: The computer architecture. The computer organization. Both (computer architecture and organization). None of the above.

By improving the input/output system of a computer, we have obtained a reduction of 20% in the execution time for a given test program. Speedup is: 25%. 20%. 50%. None of the above.