|
|
@@ -495,7 +495,11 @@ visualize the task graph more easily.
|
|
|
@section Simulated performance
|
|
|
|
|
|
StarPU can use Simgrid in order to simulate execution on an arbitrary
|
|
|
-platform. The idea is to first compile StarPU normally, and run the application,
|
|
|
+platform.
|
|
|
+
|
|
|
+@subsection Calibration
|
|
|
+
|
|
|
+The idea is to first compile StarPU normally, and run the application,
|
|
|
so as to automatically benchmark the bus and the codelets.
|
|
|
|
|
|
@smallexample
|
|
|
@@ -513,6 +517,8 @@ Note that we force to use the dmda scheduler to generate performance
|
|
|
models for the application. The application may need to be run several
|
|
|
times before the model is calibrated.
|
|
|
|
|
|
+@subsection Simulation
|
|
|
+
|
|
|
Then, recompile StarPU, passing @code{--enable-simgrid} to @code{./configure}, and re-run the
|
|
|
application, specifying the requested number of devices:
|
|
|
|
|
|
@@ -556,3 +562,35 @@ Note: of course, if the application uses @code{gettimeofday} to make its
|
|
|
performance measurements, the real time will be used, which will be bogus. To
|
|
|
get the simulated time, it has to use @code{starpu_timing_now} which returns the
|
|
|
virtual timestamp in ms.
|
|
|
+
|
|
|
+@subsection Simulation on another machine
|
|
|
+
|
|
|
+The simgrid support even permits to perform simulations on another machine, your
|
|
|
+desktop, typically. To achieve this, one still needs to perform the Calibration
|
|
|
+step on the actual machine to be simulated, then copy them to your desktop
|
|
|
+machine (the @code{$STARPU_HOME/.starpu} directory). One can then performa the
|
|
|
+Simulation step on the desktop machine, by setting the @code{STARPU_HOSTNAME}
|
|
|
+environment variable to the name of the actual machine, to make StarPU use the
|
|
|
+performance models of the simulated machine even on the desktop machine.
|
|
|
+
|
|
|
+If the desktop machine does not have CUDA or OpenCL, StarPU is still able to
|
|
|
+use simgrid to simulate execution with CUDA/OpenCL devices, but the application
|
|
|
+source code will probably disable the CUDA and OpenCL codelets in that
|
|
|
+case. Since during simgrid execution, the functions of the codelet are actually
|
|
|
+not called, one can use dummy functions such as the following to still permit
|
|
|
+CUDA or OpenCL execution:
|
|
|
+
|
|
|
+@smallexample
|
|
|
+static struct starpu_codelet cl11 =
|
|
|
+{
|
|
|
+ .cpu_funcs = {chol_cpu_codelet_update_u11, NULL},
|
|
|
+#ifdef STARPU_USE_CUDA
|
|
|
+ .cuda_funcs = {chol_cublas_codelet_update_u11, NULL},
|
|
|
+#elif defined(STARPU_SIMGRID)
|
|
|
+ .cuda_funcs = {(void*)1, NULL},
|
|
|
+#endif
|
|
|
+ .nbuffers = 1,
|
|
|
+ .modes = {STARPU_RW},
|
|
|
+ .model = &chol_model_11
|
|
|
+};
|
|
|
+@end smallexample
|
