[BUG] Rmat C++ tests fail with RNG PC generator type #2266
Labels
Milestone
Comments
MatthiasKohl
added
? - Needs Triage
rapids-bot bot
pushed a commit
to rapidsai/raft
that referenced
this issue
May 26, 2022
RAFT's RNG class provides methods that fill a buffer with random numbers belonging to various probability distribution functions. For example`uniform`, `normal` etc. In this methods, multiple instances of random number generator are used in parallel. Each cuda thread gets it own instance of generator. The instance is initialized with three values `seed`, `subsequence` and `offset`. The value for `seed` is common for all threads and thread id is used for `subsequence`. The `offset` is kept `0` for all instances. To fill the buffer, the threads work in grid strided fashion demonstrated by loop below:
```
for (size_t i = tid; i < buffer_length; i += total_number_of_threads) {
buffer[i] = rng.next();
}
```
Due to grid-striding of loops, consecutive elements in the buffer are `ith` element in consecutive subsequences. For example, 10th and 11th elements in the buffer would be 0th element in 10th and 11th subsequences. In the unit tests for cugraph, this scheme seems to introduce a slight bias in certain cases. Easy fix to the issue is to break the lock step increment of individual subsequences.
This PR sets different offset value for each subsequence by setting `offset = subsequence`. Note that this change has no effect on the period of the random number generator.
This should fix the cuGraph issue rapidsai/cugraph#2266 for now.
Authors:
- Vinay Deshpande (https://github.com/vinaydes)
Approvers:
- Thejaswi. N. S (https://github.com/teju85)
URL: #690
rapids-bot bot
pushed a commit
that referenced
this issue
Jun 21, 2022
With the RAFT changes here: rapidsai/raft#690 we should be able to use the PC generator again. The PC generator is significantly faster. Closes #2266 Authors: - Chuck Hastings (https://github.com/ChuckHastings) Approvers: - Corey J. Nolet (https://github.com/cjnolet) - Seunghwa Kang (https://github.com/seunghwak) URL: #2356
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
Rmat C++ tests fail when using RAFT PC generator type
Steps/Code to reproduce bug
Switch RngState to raft::random::GeneratorType::GenPC here: https://github.com/rapidsai/cugraph/blob/branch-22.06/cpp/src/detail/utility_wrappers.cu#L34
Expected behavior
All unit tests should pass
Environment overview (please complete the following information)
Environment details
See CI
The text was updated successfully, but these errors were encountered: