Advanced Hotspots analysis is a fast and easy way to identify performance-critical code sections (hotspots). The periodic instruction pointer sampling performed by Intel® VTune™ Amplifier identifies code locations where an application spends more time than in others. A function may consume much time either because its code is slow or because the function is frequently called. But any improvements in the speed of such functions should have a bigger impact on overall application performance.
VTune Amplifier creates a list of functions in your application ordered by the amount of time spent in each function. By default, Advanced Hotspots analysis does not capture the function call stacks as the hotspots are collected, but it can be used to sample all processes on the system. This type of analysis uses event-based sampling collection and analyzes all the processes running on your system at the moment, providing CPU time data on whole system performance.
You still can analyze stacks for your application modules by selecting the collection level that includes stack analysis in the Advanced Hotspots pane. For example, selecting the Hotspots, call counts, stacks and context switches collection level extends the Advanced Hotspots analysis with performance, parallelism and power consumption data attributed to execution paths.
To use the Advanced Hotspots analysis, explore:
-
Configuration options (knobs)
Configuration Options
To configure options for the Advanced Hotspots analysis:
-
Click the
New Analysis button on the
Intel® VTune™ Amplifier
toolbar.
The New Amplifier Result tab opens with the Analysis Type window active.
-
From the analysis tree on the left pane, select Algorithm Analysis > Advanced Hotspots.
The analysis configuration pane opens on the right.
-
Configure the following options:
CPU sampling interval, ms field
Specify an interval between CPU samples: from 0.01 ms to 1000 ms. By default, the sampling interval for the Advanced Hotspots analysis is 1 ms.
Collection level options
Select a level of details provided with event-based sampling collection:
Select This
To Do This
Hotspots
Identify application hotspots based on such basic hardware events as Clockticks and Instructions Retired.
Hotspots, stacks and context switches
Identify hardware hotspots, explore statistically reconstructed call flow of your program and analyze thread scheduling.
Hotspots, call counts, stacks and context switches
Identify hardware hotspots, analyze thread scheduling, explore call stacks and statistically approximated number of calls to sampled functions.
Hotspots, call counts, loop trip counts, stacks and context switches
Extend the Hotspots, call counts, stacks and context switches collection with an analysis of loop trip count statistically estimated using the hardware events.
Event mode drop-down list
Limit event-based sampling collection to USER (user events) or OS(system events) mode. By default, all event types are collected.
Analyze user tasks check box
Analyze tasks specified in your code via Task API.
Analyze GPU usage check box
Analyze GPU usage and frame rate and identify whether your application is GPU or CPU bound.
Note
You are recommended to enable stack collection to detect context switches and correlate CPU and GPU usage data.
Analyze Processor Graphics hardware events drop-down menu
Analyze performance data from Intel HD Graphics based on the predefined groups of GPU metrics.
Trace OpenCL and Intel Media SDK programs check box
Capture the execution time of OpenCL™ kernels and Intel Media SDK programs on a GPU, identify performance-critical GPU computing tasks, and analyze the performance of OpenCL kernels and Intel Media SDK programs per GPU hardware metrics. This option is supported for systems with Intel HD Graphics only.
Details button
Expand/collapse a section listing the default non-editable settings used for this analysis type. If you want to modify these settings, click the Copy button in the upper right corner. VTune Amplifier creates an editable copy of this analysis type configuration and locates it under the Custom Analysis node in the analysis tree.
Note
You may generate the command line for this configuration using the Command Line... button at the bottom.
Viewpoints
You can choose to view Advanced Hotspots analysis results in any of the following viewpoints:
|
Viewpoint |
Description |
|---|---|
|
Hardware Events |
Displays statistics of monitored hardware events: estimated count and/or the number of samples collected. Use this view to identify code regions (modules, functions, code lines, and so on) with the highest activity for an event of interest. |
|
Hardware Issues |
Helps identify where the application is not making the best use of available hardware resources. This viewpoint displays metrics derived from hardware performance counters. Hover over the highlighted metrics values in the grid to read why the extreme value might represent a performance problem. |
|
Hotspots |
Helps identify hotspots - code regions in the application that consume a lot of CPU time. |
|
Bandwidth |
Helps understand how effectively your application uses memory resources and identify potential memory access related issues like excessive access to remote memory on NUMA platforms, hitting DRAM or Intel® QuickPath Interconnect (Intel QPI) bandwidth limit, and others. It provides various performance metrics for both the application code and memory objects arrays. |
These viewpoints may include the following windows:
-
Summary window displays statistics on the overall application execution.
-
PMU Events window displays count for PMU events selected for the analysis.
-
Uncore Events window displays counts of uncore events selected for the analysis. If there are no uncore events, the upper pane of the window is empty.
-
Caller/Callee window displays parent and child functions of the selected focus function. This window is available only if stack collection was enabled during analysis configuration.
-
Top-down Tree window displays hotspot functions in the call tree, performance metrics for a function only (Self value) and for a function and its children together (Total value).
-
Platform window provides details on CPU and GPU utilization, frame rate, memory bandwidth, and user tasks (if corresponding metrics are collected).
-
Graphics window displays overtime data as GPU usage on a software queue, CPU time usage, OpenCL™ kernels data, and GPU performance per the Overview group of GPU hardware metrics, Memory Bandwidth, and CPU Frequency.
What's Next
You can go from the hotspots to the source code. View the source code containing the hotspots and modify your code to remove bottlenecks and improve the performance of your application.
Information provided by Advanced Hotspots analysis is important for tuning serial applications and it is still useful for tuning the serial sections of parallel applications. For algorithm tuning, you may also choose to run the Basic Hotspots analysis and analyze the call flow of the application or run the Concurrency analysis to estimate the effectiveness of the parallel algorithms you use. For Intel® Xeon Phi™ coprocessor analysis, you may run the General Exploration analysis with additional metrics that help triage hardware issues in programs running on the coprocessor.