GPU Programming

Which is not an OpenCL Architecture?

• The Platform Model
• The Programming Model
• The Execution Model
• The Kernel Model

What is a high performance multi-core processor that can be used to accelerate a wide variety of applications using parallel computing.

• CPU
• DSP
• GPU
• CLU

CUDA is a parallel computing platform and programming model 

• True
• False

The OpenCL memory model defines how data is stored and communicated both within a device and also between a device and the ________

• GPU
• Host CPU
• Constant Memory
• Kernel

CUDA is limited to Nvidia GPU only

• True
• False

OpenCL compiler is typically invoked at 

• Compile Time
• Runtime

Available to any work-item for both read and write access. 

• Private Memory
• Local Memory
• Constant Memory
• Global Memory

Which is not one of the five data structures?

• cl_device_id
• cl_context_key
• cl_program
• cl_kernel

Identifies the work-item among all other work-items executing the kernel. 

• Global ID
• Local ID

The kernel executes on the host system, defines devices context, and queues kernel execution instances using command queues.

• True
• False