As the HPC industry moves toward a hybrid computing model to meet the increasing demands for higher performance, Rogue Wave is keeping pace by providing support for GPU parallel processors, specifically for CUDA, NVIDIA’s parallel computing architecture. NVIDIA calls attention to the entire software ecosystem that has developed around the CUDA architecture—software applications, libraries, programming language solutions, and service providers. Support for CUDA in Rogue Wave’s TotalView debugger is available as an add-on feature to TotalView for Linux-x86-64, and includes the following:
-
Support for CUDA 3.2 and CUDA 4.0
-
Debugging host and device code in the same session
- CUDA running directly on Tesla or Fermi hardware
- Linux and GPU device thread visibility
-
Full visibility to the hierarchical device, block, and thread memory
-
Navigating device threads by logical and device coordinates
-
CUDA function calls, host pinned memory regions and CUDA contexts
- Handling CUDA functions inline and on the stack
-
Command line interface (CLI) commands for CUDA functions
-
Applications that use multiple NVIDIA devices at the same time
-
MPI applications on CUDA-accelerated clusters
- Support for Unified Virtual Addressing and GPUDirect
- Support for CUDA C++ and inline PTX
- Reports memory errors and handles CUDA exceptions
>> Read the latest press release about Rogue Wave and NVIDIA
>> Learn more about TotalView