Model Offloading to a GPU
- Determine if you should offload your code to a target device (for code running on a CPU) or run it on a different target device (for code running on a GPU) and what is the potential speedup before getting a hardware
- Identify loops that are recommended for offloading from a baseline CPU to a target GPU
- Pinpoint potential performance bottlenecks on the target device to decide on optimization directions
- Check how effectively data can be transferred between host and target devices
- CPU-to-GPU offload modeling for C, C++, and Fortran applications: Analyze a C, C++, or Fortran application and model its performance on a target GPU device. Use this workflow to find offload opportunities and prepare your code for efficient offload to the GPU.
- CPU-to-GPU offload modeling for Data Parallel C++ (DPC++), OpenMP* target, and OpenCL™ applications: Analyze a DPC++, OpenMP target, or OpenCL applicationoffloaded to a CPUand model its performance on a target GPU device. Use this workflow to understand how you can improve performance of your application on the target GPU and check if your code has other offload opportunities. This workflow analyzes parts of your application running on host and offloaded to a CPU.
- GPU-to-GPU offload modeling for DPC++, OpenMP target, and OpenCL applications(technical preview): Analyze DPC++, OpenMP target, or OpenCL application running on a GPU and model its performance on a different GPU device. Use this workflow to understand how you can improve your application performance and check if you can get a higher speedup if you offload the application to a different GPU device.
How It Works
- Get the baseline performance data for your application by running aSurveyanalysis.
- Identify the number of times loops are invoked and executed and the number of floating-point and integer operations, estimate cache and memory traffics on target device memory subsystem by running theCharacterizationanalysis.
- Mark up loops of interest and identify loop-carried dependencies that might block parallel execution by running theDependenciesanalysis.
- Estimate the total program speedup on a target device and other performance metrics according to Amdahl's law, considering speedup from the most profitable regions by runningPerformance Modeling. A region is profitable if its execution time on the target is less than on a host.
Offload Modeling Summary
- Main metrics for the modeled performance of your program indicating if you should offload your application to a target device or not
- Specific factors that prevent your code from achieving a better performance if executed on a target device (the factors that your code is bounded by)
- Top five offloaded loops/functions that provide the highest benefit and top five non-offloaded loops/functions with why-not-offloaded reasons