Open main menu

CDOT Wiki β

Changes

GPU610/TeamEh

1,466 bytes added, 17:51, 3 October 2014
Assignment 1
Due to this, I choose Ben's image processing for parallelisation.
 
==== Balint Czunyi's Results ====
 
=== Introduction ===
 
My Project used a Heat Equation calculator.
 
Source: https://github.com/MyCodes/Heat-Equation
 
After some changes to the Makefile to work with the profiler and testing several different parameters for the calculations I have come up with the following results:
 
=== Explicit from -15 to +15 ===
 
Each sample counts as 0.01 seconds.
% cumulative self self total
time seconds seconds calls ms/call ms/call name
34.78 0.08 0.08 __fentry__
26.09 0.14 0.06 2 30.00 31.67 Heat::writePlot()
26.09 0.20 0.06 _mcount_private
8.70 0.22 0.02 8986009 0.00 0.00 std::vector<double, std::allocator<double> >::operator[](unsigned long)
4.35 0.23 0.01 1 10.00 26.66 Heat::solveEquation(char)
0.00 0.23 0.00 8986009 0.00 0.00 std::vector<std::vector<double, std::allocator<double> >, std::allocator<std::vector<double, std::allocator<double> > > >::operator[](unsigned long)
0.00 0.23 0.00 1496502 0.00 0.00 heatFunction(double, double)
.
.
.
 
=== Summary ===
 
As you can see the std::vector<double, std::allocator<double> >::operator[](unsigned long) function is where
8.7 % of the time is spent and thus this is where the program would most benefit from parallelisation.
 
===============
=== Assignment 2 ===
=== Assignment 3 ===