Running Platform Analysis from the Command Line

CPU/GPU Concurrency Analysis

You can use the amplxe-cl command tool to profile applications that do computations on an Intel® HD Graphics and Intel® Iris™ Graphics device.

Analyzing GPU activity helps you identify whether:

• your workload is CPU- or GPU-bound.

• Intel integrated graphics are used at their full potential and the headroom for improvement, if any.

The CPU/GPU Concurrency analysis type explores code execution on the various CPU and GPU cores in your system, correlates CPU and GPU activity and identifies whether your application is GPU or CPU bound.

To perform CPU/GPU Concurrency analysis from command line, use the collect action and specify the cpugpu-concurrency predefined analysis type:

$ amplxe-cl -collect cpugpu-concurrency -- <target> [target_options]

where:

• target is the application or process (for Android*) to run

• target_options are application arguments

GPU Hotspots Analysis

Use this analysis type to identify GPU tasks with high GPU utilization and estimate the effectiveness of this utilization. The tool infrastructure automatically aligns clocks across all cores in the entire system so that you can analyze some CPU-based workloads together with GPU-based workloads within a unified time domain.

Use the GPU Hotspots analysis to:

• Identify how effectively your application uses OpenCL kernels (for Linux* and Window* targets only)

• Analyze execution of Intel Media SDK tasks over time (for Linux targets only)

• Explore GPU usage and analyze a software queue for GPU engines at each moment of time

• Explore the performance of your application per selected GPU metrics over time

Use the following command line syntax to run the GPU Hotspots analysis:

$ amplxe-cl -collect gpu-hotspots [<-knob> <knob_name=knob_option>] -- <target> [target_options]

Disk Input and Output Analysis

This analysis type uses the hardware event-based sampling collection and system-wide Ftrace* collection to provide a consistent view of the storage sub-system combined with hardware events and an easy-to-use method to match user-level source code with I/O packets executed by the hardware.

The analysis actively relies on the data produced by the kernel block driver system. In case your platform utilizes a non-standard block driver sub-system (for example, user-space storage drivers), disk metrics will not be available in the analysis type.

Use the Disk Input and Output analysis to identify:

• Imbalance between I/O and compute operations (HPC applications)

• Long latency of I/O requests (transactional workloads)

• Hardware utilization (streaming)

Use the following command line syntax to run the Disk Input and Output analysis:

$ amplxe-cl -collect disk-io -- <target> [target_options]

Prerequisites:

• Run the VTune Amplifier command line interface with root privileges.
• Enable system-wide profiling:
  • For system analysis WITH a target application or process specify your application or process and enable system-wide analysis with the -analyze-system option.
  • For system analysis WITHOUT a target application or process specify the collection duration with the -duration option.

System Overview Analysis

System Overview analysis evaluates general behavior of Linux* or Android* target systems and correlates power and performance metrics with IRQ handling. This analysis type uses the Driverless Event-Based Sampling Collection.

Use the following command line syntax to run the System Overview analysis:

$ amplxe-cl -collect system-overview [<-knob> <knob_name=knob_option>] -- <target>