Let K20c use old stream management way?

Accelerated Computing CUDA CUDA Programming and Performance
1

From K20 different streams becomes fully concurrent(used to be currently on the edge).

However My program need the old way. Or I need to do a lot of synchronization to solve the dependency problem.

Is it possible to switch stream management to the old way?

2

No, it is not possible. The new way is built into the hardware itself. Streams have always been documented as fully concurrent. If you are calling many async memcpies and kernels that have dependances on each other, then they should be in the same stream.