Company Events Academic Community Support Solutions Products & Services Contact NI MyNI

Glossary (GPU Analysis Toolkit)

LabVIEW 2012 GPU Analysis Toolkit Help

Edition Date: June 2012

Part Number: 373575A-01

»View Product Info
Download Help (Windows Only)

A

allocating memory The process of acquiring a memory buffer on the device to store both the source data that you download to the device and the results of the calculations performed on the device to upload to LabVIEW. For more information about this process, refer to the Performing FFT or BLAS Operations on a GPU Device topic.

C

CUBLAS handle A pointer that LabVIEW creates when you initialize the CUBLAS library. A handle is a pointer to the master pointer maintained by the memory manager, which updates the pointer if you move the handle to a new address.
CUFFT handle A pointer that LabVIEW creates when you initialize the CUFFT library. A handle is a pointer to the master pointer maintained by the memory manager, which updates the pointer if you move the handle to a new address.

D

data pointer The physical address of a set of data on the host computer or on the GPU device. See also device data.
device data The cluster of data that contains pointers, data properties, and the raw data from the device. Device data also can refer to the raw data independent of the associated data properties and pointers.
device pointer The physical address of a device resource. A device pointer usually refers to the location of a memory buffer.
device pointer class A LabVIEW class to store device pointer data.
device pointer data Any data that contains a device pointer.

F

freeing memory The process of releasing the memory buffer on the device that stores the device data. Refer to the Free Memory VI for more information about releasing allocated memory. To end most GPU applications, you must release the memory you allocate on the device, free the initialized resources, such as the CUFFT or CUBLAS library, and free the resources for communication with a device and for active processes. For more information about this workflow, refer to the Designing the Block Diagram to Compute on a GPU Device topic.

 

Your Feedback! poor Poor  |  Excellent excellent   Yes No
 Document Quality? 
 Answered Your Question? 
Add Comments 1 2 3 4 5 submit