Intel® Advisor provides design tools to help ensure your C/C++, Fortran, Data Parallel C++, OpenMP*, and OpenCL™ native/managed applications realize full performance potential on modern processors: Vectorization and Code Insights, CPU Roofline / Memory Insights, Threading, Offload Modeling, GPU Roofline Insights, and Flow Graph Analyzer.
CPU Roofline / Memory Insights and GPU Roofline Insights features visualize actual performance against hardware-imposed performance ceilings (rooflines). They provide insights into where the bottlenecks are, which loops are worth optimizing for performance, what are the likely causes of bottlenecks and what should be the next optimization steps.
Understand the Roofline
Learn what the Roofline is and how it works using the following resources:
- Intel® Advisor User Guide: Roofline Methodology
- Article: Intel Advisor Roofline
- Video: An Introduction to the Roofline Feature in Intel® Advisor 2017
Learn how to run the first Roofline Analysis and understand the results using the following resources:
- Use a C++ sample application and follow along the CPU Roofline tutorial to make optimization decisions with the Roofline chart.
- Learn how to use the CPU Roofline analysis to identify CPU performance bottlenecks in your application.
- Learn how to use the GPU Roofline to identify GPU performance bottlenecks in your application.
Explore More Roofline Capabilities
Intel Advisor provides several advance features of the Roofline to help you visualize actual performance against hardware-imposed performance ceilings, as well as determine the main limiting factor (memory bandwidth or compute capacity), thereby providing an ideal roadmap of potential optimization steps.
Start with the User Guide, which provides an overview of all Roofline analysis capabilities available in the Intel Advisor:
To explore a specific Roofline feature, see a corresponding page:
- Use Roofline with Callstacks to get total data metrics about your applications and investigate the source of loop/function in the entire call chain.
- Use Integer Roofline to analyze applications with integers and generate roofs with respect to integer-based scalar and vector computation peaks.
- Use Memory-Level CPU Roofline or Memory-Level GPU Roofline to analyze integrated traffic data for your application an all memory levels.
Roofline Use Cases
Intel Advisor Cookbook contains step-by-step instructions to guide you through specific Roofline use-cases:
- Identify Bottlenecks Iteratively: Cache-Aware Roofline
- Visualize Performance Improvements with CPU Roofline Compare or GPU Roofline Compare
- Identify Code Regions to Offload to GPU and Visualize GPU Usage with GPU Roofline
- Use Intel® Advisor Command Line Interface to Analyze a GPU Application
- Optimize Memory Access Patterns using Loop Interchange and Cache Blocking Techniques with the help of CPU Roofline
Notices and Disclaimers
Intel technologies may require enabled hardware, software or service activation.
No product or component can be absolutely secure.
Your costs and results may vary.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.
No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.
The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.
Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.