User Guide


CPU Roofline Accuracy Presets

For each perspective, you can choose collection accuracy level. Each accuracy level is a preset of analysis stages and properties with a corresponding accuracy value. The higher accuracy value you choose, the higher runtime overhead is added to the application you run with
Intel® Advisor
The following accuracy levels are available:
Runtime Overhead
Analyses and Properties
1.1 - 5x longer
Survey + Characterization (FLOP)
Cache-Aware Roofline Model report:
  • Plotted for L1 cache hierarchy level
  • Loop represented by a single dot
  • Metrics related only to the loop/function itself excluding data originating in other loops/functions called by it (self data)
10 - 50x longer
Survey + Characterization (Trip Counts and FLOP, Call Stacks, Memory-Level) + Memory Access Patterns
Memory-Level Roofline Model report with call stacks:
  • Plotted for all memory levels (L1, L2, L3, DRAM)
  • Loop is represented by multiple dots for different memory levels
  • Data reported for the loop/function itself and its inner loops/functions (total data)
Allows you to examine a selected loop/function in greater detail, displaying labeled dots with arithmetic intensity for the loop/function at each memory level.
You can choose custom accuracy and set a custom perspective flow for your application. For more information, see Customize
CPU / Memory Roofline Insights
There is a variety of techniques available to minimize data collection, result size, and execution overhead. Check Minimize Analysis Overhead.

Product and Performance Information


Performance varies by use, configuration and other factors. Learn more at