How many steps are needed to complete the following vector addition instruction
ID: 3858924 • Letter: H
Question
How many steps are needed to complete the following vector addition instruction on a vector processing machine whose CPU (which has a pipeline structure) is composed of 10 functional units? z(1: 100) = x(1: 100) + y(1: 100) 109 110 10 100 What is the primary reason for using parallel computing? We cannot rely on increasing the speed of CPU to meet the needs for more computational power Parallel computing is another programming paradigm Parallel computing is a natural way of programming Hardware technology makes building supercomputers feasible A multiprocessor consists of 100 processors. If 10% of the code is sequential and 90% is parallelizable, what will be the maximum speed up when running this program on this multiprocessor? ~9.17 ~100.00 ~90.00 ~10.00 Operations in SIMD machines are synchronized automatically True FalseExplanation / Answer
15. Answer : 109
Explanation :
If the pipeline has 10 functional units, each instruction will take 10 clock cycles to complete, x(1:00) + y(1:100) need 100 instructions to complete.
Each instruction will start after 1st clock cycle of previous instruction is completed.
1st instruction will start at 1st clock cycle
2nd instruction will start at 2nd clock cycle
3rd instruction will start at 3rd clock cycle
and so on...
100th instruction will start at 100th clock cycle
100 th instruction will take 10 cycles to complete,i.e. 9 more cycles after 100th clock cycle.
Thus, total Clock cycles : 100 + 9 = 109
16. Parallel programming is required because we cant increase the speed of a CPU to meet the needs of more computational power, instead we can harvest multiple CPU and their computational powers to meet the needs.
Answer : A
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.