When scaling your applications for optimal performance on multiple cores or multiples architectures, you’ll likely hit some roadblocks. This session demonstrates how to find them and remove them.
Are you finding roadblocks on the path to scaling your application to more cores? Or perhaps you can’t seem to keep all of your threads busy?
If so, this session is for you.
Learn how to liberate your application from lock and synchronization problems using Intel® VTune™ Profiler, a performance tuning and analysis tool targeting CPU, GPU, and FPGA systems.
- Identify and resolve threading synchronization issues
- Increase application performance and scaling across one or multiple platforms
- Use VTune in combination with the Intel® MPI Library and Intel® C++ Compiler Classic for application compilation and at runtime
You’ll be provided with the application code so you can follow along during the session or experiment on your own.