Difference between revisions of "Dev/null"
Ozal Islam (talk | contribs) (→Headline text) |
Ozal Islam (talk | contribs) (→Precision of Pi) |
||
| Line 21: | Line 21: | ||
Test runs with number of iterations: | Test runs with number of iterations: | ||
| + | <pre> | ||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100 | ||
| − | |||
Compute pi - took - 0 millisecs | Compute pi - took - 0 millisecs | ||
| − | |||
3.131592903558551910236928961239755153656005859375 | 3.131592903558551910236928961239755153656005859375 | ||
real 0m0.137s | real 0m0.137s | ||
| − | |||
user 0m0.004s | user 0m0.004s | ||
| − | |||
sys 0m0.008s | sys 0m0.008s | ||
| − | |||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 5000 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 5000 | ||
| − | |||
Compute pi - took - 0 millisecs | Compute pi - took - 0 millisecs | ||
| − | |||
3.14139265359178754266622490831650793552398681640625 | 3.14139265359178754266622490831650793552398681640625 | ||
real 0m0.145s | real 0m0.145s | ||
| − | |||
user 0m0.004s | user 0m0.004s | ||
| − | |||
sys 0m0.004s | sys 0m0.004s | ||
| − | |||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 10000 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 10000 | ||
| − | |||
Compute pi - took - 1 millisecs | Compute pi - took - 1 millisecs | ||
| − | |||
3.141492653590030936783250581356696784496307373046875 | 3.141492653590030936783250581356696784496307373046875 | ||
real 0m0.143s | real 0m0.143s | ||
| − | |||
user 0m0.000s | user 0m0.000s | ||
| − | |||
sys 0m0.012s | sys 0m0.012s | ||
| − | |||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100000 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100000 | ||
| − | |||
Compute pi - took - 12 millisecs | Compute pi - took - 12 millisecs | ||
| − | |||
3.14158265358971622305261917063035070896148681640625 | 3.14158265358971622305261917063035070896148681640625 | ||
real 0m0.141s | real 0m0.141s | ||
| − | |||
user 0m0.012s | user 0m0.012s | ||
| − | |||
sys 0m0.012s | sys 0m0.012s | ||
| − | |||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000 | ||
| − | |||
Compute pi - took - 139 millisecs | Compute pi - took - 139 millisecs | ||
| − | |||
3.141591653589779209454491137876175343990325927734375 | 3.141591653589779209454491137876175343990325927734375 | ||
real 0m0.269s | real 0m0.269s | ||
| − | |||
user 0m0.140s | user 0m0.140s | ||
| − | |||
sys 0m0.008s | sys 0m0.008s | ||
| − | |||
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000000 | oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000000 | ||
| − | |||
Compute pi - took - 178576 millisecs | Compute pi - took - 178576 millisecs | ||
| − | |||
3.14159265258789854868837210233323276042938232421875 | 3.14159265258789854868837210233323276042938232421875 | ||
real 2m58.717s | real 2m58.717s | ||
| − | |||
user 2m58.315s | user 2m58.315s | ||
| − | |||
sys 0m0.040s | sys 0m0.040s | ||
| − | + | </pre> | |
| − | |||
Sample GProf profile: | Sample GProf profile: | ||
| − | + | <pre>Flat profile: | |
| − | Flat profile: | ||
Each sample counts as 0.01 seconds. | Each sample counts as 0.01 seconds. | ||
| − | |||
% cumulative self self total | % cumulative self self total | ||
| − | |||
time seconds seconds calls Ts/call Ts/call name | time seconds seconds calls Ts/call Ts/call name | ||
| − | |||
100.00 10.05 10.05 compute_pi(int) | 100.00 10.05 10.05 compute_pi(int) | ||
| − | + | 0.00 10.05 0.00 1 0.00 0.00 _GLOBAL__sub_I__Z10compute_pii</pre> | |
| − | 0.00 10.05 0.00 1 0.00 0.00 _GLOBAL__sub_I__Z10compute_pii | ||
=== Assignment 2 === | === Assignment 2 === | ||
=== Assignment 3 === | === Assignment 3 === | ||
Revision as of 18:36, 14 October 2015
GPU610/DPS915 | Student List | Group and Project Index | Student Resources | Glossary
Contents
/dev/null
Team Members
- Yehoshua Ghitis, Stuff
- Boris Bershadsky, Prime generator
- Ozal Islam, Precision of Pi
oaislam1@myseneca.ca, bbershadsky@myseneca.ca?subject=dps901-gpu610 Email All
Progress
Assignment 1
Precision of Pi
The number of miliseconds required to complete the function for computing the precision of Pi takes longer if the number of iterations is increased by 10 exponentially.
If GPU was used to compute the function then the number of miliseconds could be drastically reduced with high number of iterations.
Test runs with number of iterations:
oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100 Compute pi - took - 0 millisecs 3.131592903558551910236928961239755153656005859375 real 0m0.137s user 0m0.004s sys 0m0.008s oaislam1@matrix:~/Ozal/GPU610/a1> time a1 5000 Compute pi - took - 0 millisecs 3.14139265359178754266622490831650793552398681640625 real 0m0.145s user 0m0.004s sys 0m0.004s oaislam1@matrix:~/Ozal/GPU610/a1> time a1 10000 Compute pi - took - 1 millisecs 3.141492653590030936783250581356696784496307373046875 real 0m0.143s user 0m0.000s sys 0m0.012s oaislam1@matrix:~/Ozal/GPU610/a1> time a1 100000 Compute pi - took - 12 millisecs 3.14158265358971622305261917063035070896148681640625 real 0m0.141s user 0m0.012s sys 0m0.012s oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000 Compute pi - took - 139 millisecs 3.141591653589779209454491137876175343990325927734375 real 0m0.269s user 0m0.140s sys 0m0.008s oaislam1@matrix:~/Ozal/GPU610/a1> time a1 1000000000 Compute pi - took - 178576 millisecs 3.14159265258789854868837210233323276042938232421875 real 2m58.717s user 2m58.315s sys 0m0.040s
Sample GProf profile:
Flat profile: Each sample counts as 0.01 seconds. % cumulative self self total time seconds seconds calls Ts/call Ts/call name 100.00 10.05 10.05 compute_pi(int) 0.00 10.05 0.00 1 0.00 0.00 _GLOBAL__sub_I__Z10compute_pii
