Expanded Accelerator Profiling
- Identify occupancy issues that prevent GPU code from efficiently using available execution unit (EU) threads by using the GPU Compute/Media hot spots analysis.
- Analyze data transfer to pinpoint inefficient code paths between host and device in the GPU Offload Analysis.
- Profile DirectX applications on the CPU host to determine the gaps between the API calls and the reasons causing such inefficiencies.
- Get support for the 3rd generation Intel® Xeon® Scalable processors (formerly code named Ice Lake server) and 12th generation Intel® Core™ processor family (formerly code named Alder Lake)
- Visualize hot code paths and identify time spent in each function and its callees using the Flame Graph in Hot Spots analysis.
- Diagnose the issues that may cause your system to be overheated or consume significant power using the CPU Throttling analysis.
- Gain insight into system configuration, performance, and behavior with Platform Profiler—now fully integrated into Intel VTune Profiler.
- Quickly analyze the application performance characterization with pointers to deeper analysis using Performance Snapshot.
- Expanded I/O Analysis: Quickly visualize I/O hardware use with a platform diagram. Discover the root cause of low throughput in Intel Data Direct I/O Technology. Optimize latency using extended utilization metrics in I/O analysis. Get added support for MPI applications.
- Application Performance Snapshot: Analyze workloads at scale to identify outliers and where they occurred. Get PCIe bandwidth information in command-line reports.
For a more complete and up-to-date list, see the release notes.
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.