# Calculating GFLOPS (fft, LU, sort...) How to calculate GFLOPS for various problems?

**URL:** <https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669>\
**Category:** CUDA Programming and Performance\
**Created:** [January 31, 2010, 8:04pm UTC](https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669 "2010-01-31T20:04:54Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![DL85](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@DL85](https://forums.developer.nvidia.com/u/DL85)\
**Post date:** [January 31, 2010, 8:04pm UTC](https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669/1 "2010-01-31T20:04:54Z")

</div>

Hi!  
I’m to write a benchmark suite for CUDA, including fft, LU decomposition, sparse solver and bitonic sort. I’m wondering where all these GFLOP counts come from? How do i measure GFLOPS for each of the tasks? Do I have to know the implementation of those algorithms to calculate it?

thanks  
Dell

---

<div class="post-metadata">

**Author:** ![avidday](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/avidday/32/59219_2.png) [@avidday](https://forums.developer.nvidia.com/u/avidday)\
**Post date:** [January 31, 2010, 8:27pm UTC](https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669/2 "2010-01-31T20:27:40Z")

</div>

Many of those functions you mention have known (or at least bounded) theoretical algorithmic complexity. That is how classic benchmarks like the LINPACK benchmark work - the FLOP value is based purely on the theoretical operation count and the execution time. The actual implementation is not considered.

---

<div class="post-metadata">

**Author:** ![Nikolai](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Nikolai](https://forums.developer.nvidia.com/u/Nikolai)\
**Post date:** [January 31, 2010, 10:25pm UTC](https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669/3 "2010-01-31T22:25:42Z")

</div>

may i ask which sparse solver you were planning on benchmarking?

---

<div class="post-metadata">

**Author:** ![DL85](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@DL85](https://forums.developer.nvidia.com/u/DL85)\
**Post date:** [January 31, 2010, 10:50pm UTC](https://forums.developer.nvidia.com/t/calculating-gflops-fft-lu-sort-how-to-calculate-gflops-for-various-problems/14669/4 "2010-01-31T22:50:37Z")

</div>

here is one: [url=“[Iterative CUDA](http://mathema.tician.de/software/iterative-cuda)”][http://mathema.tician.de/software/iterative-cuda[/url]](http://mathema.tician.de/software/iterative-cuda%5B/url%5D), and also CNC solver