Interface | Description |
---|---|
GPUMemoryAllocator |
Class | Description |
---|---|
CSRPointer |
Compressed Sparse Row (CSR) format for CUDA
Generalized matrix multiply is implemented for CSR format in the cuSparse library among other operations
Since we assume that the matrix is stored with zero-based indexing (i.e.
|
CudaMemoryAllocator | |
ExecutionConfig |
Java Wrapper to specify CUDA execution configuration for launching custom kernels
|
GPUContext |
Represents a context per GPU accessible through the same JVM.
|
GPUContextPool | |
GPULazyCudaFreeMemoryManager | |
GPUMatrixMemoryManager | |
GPUMemoryManager |
- All cudaFree and cudaMalloc in SystemDS should go through this class to avoid OOM or incorrect results.
|
GPUMemoryManager.EvictionPolicyBasedComparator |
Class that governs the eviction policy
|
GPUObject |
Handle to a matrix block on the GPU
|
JCudaKernels | |
ShadowBuffer | |
UnifiedMemoryAllocator |
Copyright © 2020 The Apache Software Foundation. All rights reserved.