Sample: batchCUBLAS
Minimum spec: SM 1.1

A CUDA Sample that demonstrates how using batched CUBLAS API calls to improve overall performance.

Key concepts:
