Define: an acronym for Compute Unified Device Architecture

Advantages over GPGPU:

Scattered reads – code can read from arbitrary addresses in memory.

Shared memory – CUDA exposes a fast shared memory region (up to 48KB in size) that can be shared amongst threads. 

Faster downloads and readbacks to and from the GPU

Full support for integer and bitwise operations, including integer texture lookups.

Limitations:

Fermi GPUs(compute capability 2.0) have(nearly) full support of C++, but the member functions can't be virtual.

Texture rendering is not supported.(we don't concern!)

Double precision only supports round-to-nearest-even and chop operations.

The bus bandwidth and latency between the CPU and the GPU may be a bottleneck.

Threads should be running in groups of at least 32 for best performance, with total number of threads numbering in the thousands. Branches in the program code do not impact performance significantly, provided that each of 32 threads takes         the same execution path.