Operating Models
Learn how to use Intel® VTune™ Profiler in operating models that cater to various workflows.
- Install VTune Profiler on Supported Platforms
- Use VTune Profiler as a Server (Web Browser)
- Use VTune Profiler in a Cloud Environment
- Use VTune Profiler in a Container
- Install VTune Profiler on a Virtual Machine
- Integrate VTune Profiler into your Continuous Integration (CI) Pipeline
- Profile with a Target Package
Install VTune Profiler on Supported Platforms
Install VTune Profiler on these platforms:
You can install VTune Profiler :
- As a standalone package
- As a component in the Intel® oneAPI Base Toolkit
Learn More:
- Release Notes about features introduced in versions of VTune Profiler
- Technical compatibility information in System Requirements
Use VTune Profiler as a Server (Web Browser)
Purpose:
Run VTune Profiler as a web service. Access the application and run profiling analyses through a web browser.
Benefits:
- Access VTune Profiler in remote test environments.
- Run VTune Profiler on the target or remote development machine(s) without having to install the standalone package on any of them.
- Manage fewer system requirements in comparison to the standalone package.
- Share collection results through a URL.
- Deploy into your web hosting infrastructure or as a standalone service. VTune Profiler Server supports integration with identity providers through SAML SSO.
- Profile a pool of targets by sharing data collection agents.
Learn More:
Use VTune Profiler In a Cloud Environment
Purpose:
Install VTune Profiler in public cloud instances from these Cloud Service Providers (CSPs):
- Amazon Web Service* (AWS)
- Google Cloud Platform*
- Microsoft Azure*
Benefits:
- Run user-mode analyses (Hotspots and Threading) in virtualized environments.
- Run hardware event-based analysis types on bare-metal instances.
Learn More:
Use VTune Profiler In a Container
Purpose:
Use VTune Profiler in popular containerized environments like Docker*.
Benefits:
- Run VTune Profiler inside the container.
- Analyze containerized workloads.
- Analyze a workload within the same container.
- Analyze non-containerized workloads while running inside container.
- Analyze multiple containers simultaneously.
Learn More:
- A technical recipe on Profiling Docker* Containers
Install VTune Profiler on a Virtual Machine
Purpose:
Install VTune Profiler on a virtual machine in the same way you would install on a bare-metal machine. Find support for these hypervisors:
To ensure maximum compatibility, you may need to complete additional configuration for each of these hypervisors.
Also, due to the inherent nature of virtualization, some limitations may exist in the way you access hardware performance counters.
Learn More:
Integrate VTune Profiler into your Continuous Integration (CI) Pipeline
Purpose:
Integrate VTune Profiler into your Continuous Integration pipeline to collect performance data in specific conditions.
Benefits:
- Track workload performance over time.
- Spot and investigate performance regressions.
- Use the command line interface (CLI) to run an analysis type, or a combination of analysis types, on your workload.
Learn More:
- A technical recipe about Enabling Performance Profiling in GitLab* CI
Profile With a Target Package
Purpose:
A target package is a minimal distribution of VTune Profiler. You use target packages to enable profiling capabilities on specific remote targets. With target packages, you do not install VTune Profiler on the target system. Instead, you copy the target package to the target system and apply additional configurations to enable remote profiling from a natively supported system.
In the case of FreeBSD* OS, you can also use VTune Profiler directly on the FreeBSD system through the command line interface.
Benefits:
- Profile remote targets without installing the complete package of VTune Profiler.
- For FreeBSD targets, run command line analyses directly.
Learn More:
Parent topic: Intel® VTune™ Profiler Installation Guide