Following are the questions on CUDA programming generally asked in interviews. Go through them.
31. Which properties of gpu are used to find its compute capability?
32. How to set any gpu as current gpu when its id is given?
33. What is the use of cudaChooseDevice() function and what arguments does it accept?
34. What is the use of memset() function and what arguments does it accept?
35. How to declare a variable which is to be saved in shared memory?
36. How to declare a variable which is to be saved in constant memory?
37. What is the use of cudaMemcpyToSymbol() function and what arguments does it accept?
38. What are the different alternatives to Cuda?
39. What is parallel computing?
40. What are the advantages of parallel computing?
41. What are the different architectures according to Flynn's taxonaomy?
42. What are different parallel programming models?
43. What are the two basic classes of parallel architectures?
44. Which are general purpose parallel architectures?
45. Which are special purpose parallel architecures?
46. What is the unit used to represent frequency of a gpu?
47. What do you understand by Mflop?
48. What do you mean by speedup with respect to performance of parallel computers?
49. What is Amdahl's law? Give the expression for the speed up.
Post a Comment