-
### 🚀 The feature, motivation and pitch
The SpGEMM algorithm in cuda 11.x version requires high amount of memory for the sparse computation. In CUDA 12, two new SpGEMM algorithms has been introduced …
-
Hello!
When using the latest JCuda to do a sparse matrix multiplication, the parameters externalBuffer1 of cusparseSpGEMM_workEstimation(...) and externalBuffer2 of cusparseSpGEMM_compute(...) are …
-
From a biological standpoint, the computational bottleneck highlights the challenge of dealing with vast amounts of genetic data to predict complex traits.
Section 2.1 of [this paper](https://www.nc…
hrluo updated
4 months ago
-
Following the guidelines in https://github.com/SparseRooflineBenchmark/SparseRooflineBenchmark/issues/16, let's create some smaller issues surrounding problem set generation. The following notes may b…
-
Reproducer that causes hang:
```
module purge
module load cmake/3.17.0 gcc/10.2.0 armpl/21.1.0
export OMP_NUM_THREADS=47
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --ar…
-
Testing with the Sycl backend on Intel Ponte Vecchio on the new Blake showed a couple failing sub-tests (failure output listed below the failing executable), depending on which environment variables s…
-
Hello!
I am running sympack on shared memory. When I use one upcxx process it is working fine.
_upcxx-run -n 1 -- /home/abdulfe/sympack/symPACK/build/run_sympack2D -in simulated_matrices/output…
-
Ont his version of the code: https://github.com/SuperScientificSoftwareLaboratory/TileSpGEMM/pull/2
and (at least) on the following matrices from https://sparse.tamu.edu/Williams:
* cant/cant.mtx
…
-
-
Computing A^2 (where a is the 494_bus matrix from suitesparse, 494x494 with 1080 nonzeros) hangs on the SYCL backend.