Starting up CUDA context...
Allocating and initializing host arrays...

Allocating and initializing CUDA arrays...

Running GPU bitonic sort (1 identical iterations)...

Testing array length 64 (16 arrays per batch)...
Average time: 10.609000 ms


Validating the results...
...reading back GPU results
...inspecting keys array: OK
...inspecting keys and values array: OK
...stability property: NOT stable

Testing array length 128 (8 arrays per batch)...
Average time: 8.176000 ms


Validating the results...
...reading back GPU results
...inspecting keys array: OK
...inspecting keys and values array: OK
...stability property: stable!

Testing array length 256 (4 arrays per batch)...
Average time: 4.874000 ms


Validating the results...
...reading back GPU results
...inspecting keys array: OK
...inspecting keys and values array: OK
...stability property: NOT stable

Testing array length 512 (2 arrays per batch)...
Average time: 4.817000 ms


Validating the results...
...reading back GPU results
...inspecting keys array: OK
...inspecting keys and values array: OK
...stability property: NOT stable

Testing array length 1024 (1 arrays per batch)...
Average time: 4.660000 ms

sortingNetworks-bitonic, Throughput = 0.2197 MElements/s, Time = 0.00466 s, Size = 1024 elements, NumDevsUsed = 1, Workgroup = 512

Validating the results...
...reading back GPU results
...inspecting keys array: OK
...inspecting keys and values array: OK
...stability property: NOT stable

Shutting down...
