The Intel® Advisor is a tool to help design and optimize high-performing code for modern computer architectures. Each chapter in the Intel® Advisor Cookbook contains step-by-step instructions to help effectively use more cores, vectorization, or heterogeneous processing using Intel Advisor.
Use Intel® Advisor Beta to optimize your application's GPU usage and increase performance. This recipe uses the GPU Roofline analysis and Offload Advisor features of the Intel® Advisor Beta to identify bottlenecks in your application and regions to offload to the GPU.
Improving an application’s performance is often a multi-step process. You can take advantage of the Intel® Advisor Cache-Aware Roofline feature and supporting analysis types to perform step-by-step, systematic optimization: identify and address bottlenecks, then re-run analyses to see how your code improves with each iteration and what to do next. This section provides an...
Optimize Memory Access Patterns using Loop Interchange and Cache Blocking Techniques de Intel® Advisor Cookbook
Understanding how your application accesses data, and how that data fits in the cache, is a critical part of optimizing the performance of your code. In this recipe, use the Memory Access Patterns analysis and recommendations to identify and address common memory bottlenecks, using techniques like loop interchange and cache blocking.
Use the Roofline Compare feature to identify similar loops or functions in different Roofline analysis results and help make informed optimization choices about your code. This section describes how to compare two Roofline analysis results to visualize improvements made by loops and functions in an application.
You can analyze application performance using the Intel® Advisor in an Amazon Web Services* (AWS*) EC2* instance. This section describes how to set up and connect to your instance, and get started using the Intel Advisor GUI or command line interface (CLI).
This section of the Intel® DPC++ Compatibility Tool User Guide supports your migration of CUDA* applications to Data Parallel C++ (DPC++) code. It complements the Intel® DPC++ Compatibility Tool. This section covers the general workflow of the migration process for an entire project, assuming the project is built with Make*/CMake* or Eclipse*.