diff --git a/CMakeLists.txt b/CMakeLists.txt
index fbb1d91d756dcf2581de3af36655c8f821350258..504725f3fd21afdcd38af5079ff7e3e87cbdf940 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1044,12 +1044,8 @@ if ( WALBERLA_BUILD_WITH_CUDA )
       find_library(CUDART_LIBRARY cudart ${CMAKE_CUDA_IMPLICIT_LINK_DIRECTORIES})
       list ( APPEND SERVICE_LIBS ${CUDART_LIBRARY} )
 
-      #TODO: check if this is really needed? is it possible that nvtx is missing?
       find_library( NVTX_LIBRARY nvToolsExt ${CMAKE_CUDA_IMPLICIT_LINK_DIRECTORIES} )
-      if( NVTX_LIBRARY )
-          set( WALBERLA_CUDA_NVTX_AVAILABLE 1)
-          list ( APPEND SERVICE_LIBS ${NVTX_LIBRARY} )
-      endif()
+      list ( APPEND SERVICE_LIBS ${NVTX_LIBRARY} )
 
       #CUDA_FOUND is need for our cmake mechanism
       set ( CUDA_FOUND TRUE )
diff --git a/apps/benchmarks/PhaseFieldAllenCahn/benchmark_multiphase.cpp b/apps/benchmarks/PhaseFieldAllenCahn/benchmark_multiphase.cpp
index 3513c15f2f13250e9462c84af59b3ef3462e21a9..8efb4756d56940111c46ca684fcb3245e4c2210d 100644
--- a/apps/benchmarks/PhaseFieldAllenCahn/benchmark_multiphase.cpp
+++ b/apps/benchmarks/PhaseFieldAllenCahn/benchmark_multiphase.cpp
@@ -49,7 +49,6 @@
 #   include "cuda/AddGPUFieldToStorage.h"
 #   include "cuda/DeviceSelectMPI.h"
 #   include "cuda/HostFieldAllocator.h"
-#   include "cuda/NVTX.h"
 #   include "cuda/ParallelStreams.h"
 #   include "cuda/communication/GPUPackInfo.h"
 #   include "cuda/communication/MemcpyPackInfo.h"
diff --git a/apps/showcases/PhaseFieldAllenCahn/GPU/multiphase.cpp b/apps/showcases/PhaseFieldAllenCahn/GPU/multiphase.cpp
index 2080ce38cf767c26b2e83ca36d64de5b5ec77cef..de54dfe91b39cce943fa15ea1338e0131842866c 100644
--- a/apps/showcases/PhaseFieldAllenCahn/GPU/multiphase.cpp
+++ b/apps/showcases/PhaseFieldAllenCahn/GPU/multiphase.cpp
@@ -27,7 +27,6 @@
 
 #include "cuda/AddGPUFieldToStorage.h"
 #include "cuda/DeviceSelectMPI.h"
-#include "cuda/NVTX.h"
 #include "cuda/ParallelStreams.h"
 #include "cuda/communication/UniformGPUScheme.h"
 
diff --git a/apps/tutorials/codegen/03_AdvancedLBMCodegen.cpp b/apps/tutorials/codegen/03_AdvancedLBMCodegen.cpp
index b6ff21d73c085938ea896d3432b0be17cf4164d7..492d40b8bf07458e2b8b847d36dff28e9356d72a 100644
--- a/apps/tutorials/codegen/03_AdvancedLBMCodegen.cpp
+++ b/apps/tutorials/codegen/03_AdvancedLBMCodegen.cpp
@@ -26,7 +26,6 @@
 #   include "cuda/AddGPUFieldToStorage.h"
 #   include "cuda/DeviceSelectMPI.h"
 #   include "cuda/HostFieldAllocator.h"
-#   include "cuda/NVTX.h"
 #   include "cuda/ParallelStreams.h"
 #   include "cuda/communication/GPUPackInfo.h"
 #   include "cuda/communication/UniformGPUScheme.h"
diff --git a/src/cuda/NVTX.h b/src/cuda/NVTX.h
index 3943581afcb7d56076bfac683be6e7aa049f7038..a8c1210b827b89d28f5d1491a84adaa9f020432e 100644
--- a/src/cuda/NVTX.h
+++ b/src/cuda/NVTX.h
@@ -23,7 +23,6 @@
 
 #include <string>
 
-#ifdef WALBERLA_CUDA_NVTX_AVAILABLE
 #include <nvToolsExt.h>
 #include <nvToolsExtCuda.h>
 #include <nvToolsExtCudaRt.h>
@@ -73,25 +72,4 @@ private:
 
 
 } // namespace cuda
-} // namespace walberla
-
-
-
-
-#else
-namespace walberla{
-namespace cuda {
-
-inline void nameStream(const cudaStream_t & stream, const std::string & name) {}
-inline void nvtxMarker(const std::string& name, const uint32_t color=0xaaaaaa) {}
-class NvtxRange
-{
-public:
-    NvtxRange(const std::string & name, const uint32_t color=0xaaaaaa) {}}
-};
-
-} // namespace cuda
-} // namespace walberla
-
-
-#endif
\ No newline at end of file
+} // namespace walberla
\ No newline at end of file
diff --git a/src/waLBerlaDefinitions.in.h b/src/waLBerlaDefinitions.in.h
index c729cd2afd5e085be341517a3442d1d7e469273b..452d6d8bad7fac8a8efa699f1f4d9b94d7bb8161 100644
--- a/src/waLBerlaDefinitions.in.h
+++ b/src/waLBerlaDefinitions.in.h
@@ -33,7 +33,6 @@
 #cmakedefine WALBERLA_MESAPD_CONVEX_POLYHEDRON_AVAILABLE
 
 #cmakedefine WALBERLA_BUILD_WITH_CUDA
-#cmakedefine WALBERLA_CUDA_NVTX_AVAILABLE
 
 #cmakedefine WALBERLA_BUILD_WITH_CODEGEN