Difference between revisions of "VTK Cuda Memory Management Classes"

From NAMIC Wiki
Jump to: navigation, search
Line 2: Line 2:
 
The Idea is to get a safe object oriented interface for the memory management of CUDA devices in VTK.
 
The Idea is to get a safe object oriented interface for the memory management of CUDA devices in VTK.
  
Also take a look at [[VTK Cuda Wrapper Classes]] for more information on VTK CUDA wrappings
+
(Also take a look at [[VTK Cuda Wrapper Classes]] for more information on VTK CUDA wrappings.)
 +
 
 
== Cuda Memory ==
 
== Cuda Memory ==
 
Cuda memory comes in different:
 
Cuda memory comes in different:
*'''Host Memory''': Non-Paged Memory on the host side for fast interaction with the CUDA devices. Defined in vtkCudaHostMemory.h
+
*'''Host Memory''': Non-Paged Memory on the host side for fast interaction with the CUDA devices. see ''vtkCudaHostMemory.h''
*'''Linear Memory''': Linerely arranged memory on the CUDA device. vtk
+
*'''Linear Memory''': Linerely arranged memory on the CUDA device. see ''vtkCudaMemory.h''
*'''Pitch Memory''': 2D memory arrays that are aligned in the correct fashion for a fast interaction on the cuda devices. (Arranged over the memory banks for quick access)
+
*'''Pitch Memory''': 2D memory arrays that are aligned in the correct fashion for a fast interaction on the cuda devices. (Arranged over the memory banks for quick access). see ''vtkCudaMemoryPitch.h''
*'''Array Memory''': Arrays allocated over the CUDA API.
+
*'''Array Memory''': Arrays allocated over the CUDA API. see ''vtkCudaMemoryArray.h''
*'''Texture Memory''': Not discussed here.
+
*'''Texture Memory''': Not discussed here. see [[VTK Cuda Wrapper Classes]]
  
 
= Class Diagram =
 
= Class Diagram =
 
[[Image:vtkCudaMemoryManagementClassDiagram.png]]
 
[[Image:vtkCudaMemoryManagementClassDiagram.png]]

Revision as of 20:57, 28 December 2007

Home < VTK Cuda Memory Management Classes

Objective

The Idea is to get a safe object oriented interface for the memory management of CUDA devices in VTK.

(Also take a look at VTK Cuda Wrapper Classes for more information on VTK CUDA wrappings.)

Cuda Memory

Cuda memory comes in different:

  • Host Memory: Non-Paged Memory on the host side for fast interaction with the CUDA devices. see vtkCudaHostMemory.h
  • Linear Memory: Linerely arranged memory on the CUDA device. see vtkCudaMemory.h
  • Pitch Memory: 2D memory arrays that are aligned in the correct fashion for a fast interaction on the cuda devices. (Arranged over the memory banks for quick access). see vtkCudaMemoryPitch.h
  • Array Memory: Arrays allocated over the CUDA API. see vtkCudaMemoryArray.h
  • Texture Memory: Not discussed here. see VTK Cuda Wrapper Classes

Class Diagram

VtkCudaMemoryManagementClassDiagram.png