5.14 Write a version of the inner product procedure described in Problem 5.13 that uses 6 x 1 loop unrolling. For x86-64, our measurements of the unrolled version give a CPE of 1.07 for integer data but still 3.01 for both floating-point data. A. Explain why any (scalar) version of an inner product procedure running on an Intel Core i7 Haswell processor cannot achieve a CPE less than 1.00. B. Explain why the performance for floating-point data did not improve with loop unrolling.

5.14 Write a version of the inner product procedure described in Problem 5.13 that uses 6 x 1 loop unrolling. For x86-64, our measurements of the unrolled version give a CPE of 1.07 for integer data but still 3.01 for both floating-point data. A. Explain why any (scalar) version of an inner product procedure running on an Intel Core i7 Haswell processor cannot achieve a CPE less than 1.00. B. Explain why the performance for floating-point data did not improve with loop unrolling.

Computer Networking: A Top-Down Approach (7th Edition)

7th Edition

ISBN:9780133594140

Author:James Kurose, Keith Ross

Publisher:James Kurose, Keith Ross

Chapter1: Computer Networks And The Internet

Section: Chapter Questions

Problem R1RQ: What is the difference between a host and an end system? List several different types of end...

See similar textbooks

Similar questions

Implement nibble_swap() function. Take in 4 byte integer and returns integer with swapped nibbles in first two bytes. Use a 3 bit register. The template below is for your reference. The code in the Assembly code portion needs to be completed in Assembly x86-64 #include extern "C" int nibble_swap(int var); int main(int avgc, char** avgs) { int v = nibble_swap(0xCAFEBABA); //answer: 0xACEFABAB printf("Swapped Variable is : 0%X \n", v); return 0; } ----------------Assembly Code --------------------------- ;yasm -f elf64 var.asm section .text global start : ;put ur code here ret
Example: The Problem Input File Using C programming language write a program that simulates a variant of the Tiny Machine Architecture. In this implementation memory (RAM) is split into Instruction Memory (IM) and Data Memory (DM). Your code must implement the basic instruction set architecture (ISA) of the Tiny Machine Architecture: //IN 5 //OUT 7 //STORE O //IN 5 //OUT 7 //STORE 1 //LOAD O //SUB 1 55 67 30 55 67 1 LOAD 2- ADD 3> STORE 4> SUB 5> IN 6> OUT 7> END 8> JMP 9> SKIPZ 31 10 41 30 //STORE O 67 //OUT 7 11 /LOAD 1 //OUT 7 //END 67 70 Output Specifications Each piece of the architecture must be accurately represented in your code (Instruction Register, Program Counter, Memory Address Registers, Instruction Memory, Data Memory, Memory Data Registers, and Accumulator). Data Memory will be represented by an integer array. Your Program Counter will begin pointing to the first instruction of the program. Your simulator should provide output according to the input file. Along with…
In microprocessor 8086 How to do adding of array If my array is 5 10 ,12,13,15,18,19,20 And average of array and max and min of array
#ammend this code for the inputs (x = 5.01, 4.99, 0, -2.01, -1.99) so that there is a tolerance/Leeway for the values off > #such that 4.99 prints blue and -2.01 prints orange. # cannot adjust the value to x > 4.99 instead of 5 must add new code to add a tolerance or Leeway if x > 5: print ("blue") elif x < -2: print ("orange") else: print("purple")
Suppose that we are developing a new version of the AMD Barcelona proces- sor with a 4 GHz clock rate. We have added some additional instructions to the instruction set in such a way that the number of instructions has been reduced by 15% from the values shown for each benchmark in Exercise 1.12. The execution times obtained are shown in the following table. a. Name Execution Time (seconds) Reference Time (seconds) bzip2 700 9650 SPECratio 13.7 1.13.2 [10] In general, these CPI values are larger than those obtained in previous exercises for the same benchmarks. This is due mainly to the clock rate used in both cases, 3 GHz and 4 GHz. Determine whether the increase in the CPI is similar to that of the clock rate. If they are dissimilar, why?