We will compare the performance of a vector processor, Basic Computer Science

In this problem we will compare the performance of a vector processor with a system
that contains a scalar processor and a GPU-based coprocessor. In the hybrid system,
the host processor has superior scalar performance to the GPU, so in this case all scalar
code is executed on the host processor while all vector code is executed on the GPU.
We will refer to the rst system as the vector computer and the second system as the
hybrid computer.
Assume your target application contains a kernel with an arithmetic intensity of 0.5
FLOPs per DRAM byte accessed. However, the application also has a scalar component
which must be performed before and after the kernel in order to prepare the input
vectors and output vectors, respectively.
For a sample dataset, the scalar portion of the code requires 400 ms of execution time
on both the vector processor and the host processor in the hybrid system. The kernel
reads input vectors consisting of 200 MB and has output data consisting of 100 MB.
The vector processor has a peak memory bandwidth of 30 GB/s and the GPU has a
peak memory bandwidth of 150 GB/s. The hybrid system has an additional overhead
that requires all input vectors to be transferred between the host memory and GPU
local memory before and after the kernel is invoked. The hybrid system has a DMA
bandwidth of 10 GB/s and an average latency of 10 ms.Assume that both the vector processor and GPU are both performance bound by mem-
ory bandwidth. Compute the execution time for both computers for this application
Posted Date: 11/3/2015 7:35:45 PM | Location :







Related Discussions:- We will compare the performance of a vector processor, Assignment Help, Ask Question on We will compare the performance of a vector processor, Get Answer, Expert's Help, We will compare the performance of a vector processor Discussions

Write discussion on We will compare the performance of a vector processor
Your posts are moderated
Related Questions

Classify computer systems according to capacity. How they are different from computers according to the classification of technology. Provide comparative study also.

Number data types store numeric values. They are an immutable data type, which means that changing the value of a number data type results in a newly allocated object. Number objec

create a flowchart showing average score for the 3 quizzes assume that there are 3 sections each having 5 students the only valid number to be entered is 1-100 for the quizzes shou

NUMBER SYSTEM:  We are familiar with decimal number system which uses ten distinct symbols from 0...9, and has base 10. In the decimal number system a number n 4 n 3 n 2 n 1


1. In each of the following situations, indicate whether f = O(g), or f = O(g), or both (in which case f = T(g)). Briefly explain why. (a) f(n)=10n5 +8n2,g(n)=20n4 +7n3 +300 (b) f

A rock weighs 33.6 N on Planet X and 49 N on Earth. What is g on Planet X

what is cai

How Should an Ideal Software be?  It should be easy to operate, with minimal training of nursing personnel.   Should be reliable, and thoroughly tested in various nur