Skip to main content

Android SoC benchmarking CLI for latency, loaded latency, and memory subsystem characterization

Project description

BenchmarkingCLI

A command-line tool for running latency benchmarks on Android devices. This tool automates the process of pushing benchmark binaries to connected Android devices, executing latency tests, and collecting performance metrics in CSV format.

Features

  • Latency Testing: Measure memory latency across different buffer sizes on a specific core
  • Loaded Latency Testing (v0.3+): Measure latency under concurrent bandwidth loads with configurable kernel modes and thread settings
  • CSV Export: Results are automatically parsed and exported to timestamped CSV files
  • CLI Interface: Easy-to-use command-line interface

Requirements

Host Machine

  • Python: 3.8 or higher
  • Android Debug Bridge (ADB): Version 1.0.40+
    • Linux: apt-get install android-tools-adb
    • macOS: brew install android-platform-tools
    • Windows: Download the SDK Platform Tools and add to PATH
    • Verify installation: adb version

Android Device

  • USB debugging must be enabled on the device. To enable it:
    1. Go to Settings → About Phone
    2. Tap Build Number 7 times until you see "You are now a developer"
    3. Go to Settings → Developer Options
    4. Enable USB Debugging
  • Connect the device via USB and accept the "Allow USB Debugging" prompt on the device
  • Confirm the device is detected: adb devices

Installation

Install directly from PyPI:

pip install benchmarkingcli

Usage

Latency Test

Run a basic latency benchmark on a specific core:

# Run on default core (7)
benchmarkingcli latency

# Run on a specific core
benchmarkingcli latency --core 4

Parameters:

  • --core: CPU core to run the latency test on (default: 7)
  • --output-dir: Directory to save results (default: current directory)

Output

Results are exported as CSV files with the following columns:

  • buffer_size_kb: Size of the buffer tested (in KB)
  • min_ns: Minimum latency (nanoseconds)
  • mean_ns: Mean latency (nanoseconds)
  • max_ns: Maximum latency (nanoseconds)

Example output file: latency_results_20260427_115540.csv

Loaded Latency Test (v0.3+)

Run a latency benchmark under concurrent bandwidth load:

# Run with default settings
benchmarkingcli loadedlatency

# Run with custom configuration
benchmarkingcli loadedlatency --lat-buffer-mb 128 --bw-buffer-mb 128 --latency-core 7 --bw-cores 0,1,2,3 --kernel-mode TRIAD --ilp 4 --injection-rate desc --output-dir ./results

Parameters:

  • --lat-buffer-mb: Latency buffer size in MB (default: 128)
  • --bw-buffer-mb: Bandwidth buffer size in MB (default: 128)
  • --latency-core: Core to run latency thread on (default: 7)
  • --bw-cores: Comma-separated cores for bandwidth threads (default: 0,1,2,3,4,5)
  • --kernel-mode: Kernel mode: SUM/COPY/UPDATE/TRIAD or 1/2/3/4 (default: TRIAD)
  • --ilp: Unroll factor: 1/2/4/8/16/32 (default: 4)
  • --injection-rate: Injection rate: asc or desc (default: asc)
  • --output-dir: Directory to save results (default: current directory)

Important Update (v0.4+): The --tile-delay-sweep-order parameter has been renamed to --injection-rate. Supported values are asc and desc.

GPU Loaded Latency Test (v0.4+)

Run a latency benchmark under concurrent GPU load:

# Run with default settings
benchmarkingcli gpuloadedlatency

# Run with custom configuration
benchmarkingcli gpuloadedlatency --latency-core 7 --lat-buffer-mb 128 --gpu-buffer-mb 128 --kernel-mode TRIAD --output-dir ./results

Parameters:

  • --latency-core: Core to run latency thread on (default: 7)
  • --lat-buffer-mb: Latency buffer size in MB (default: 128)
  • --gpu-buffer-mb: GPU workload buffer size in MB (default: 128)
  • --kernel-mode: Kernel mode: SUM/COPY/UPDATE/TRIAD or 1/2/3/4 (default: TRIAD)
  • --output-dir: Directory to save results (default: current directory)

Troubleshooting

"No ADB devices found"

  • Check that your Android device is connected: adb devices
  • Ensure USB debugging is enabled on the device
  • Try: adb kill-server then adb start-server

"Latency test binary not found"

  • Reinstall the package: pip install --force-reinstall benchmarkingcli

Binary push fails

  • Check device storage: adb shell df /data/local/tmp
  • Verify ADB permissions: adb shell ls -la /data/local/tmp

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

benchmarkingcli-0.4.0.tar.gz (2.1 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

benchmarkingcli-0.4.0-py3-none-any.whl (2.1 MB view details)

Uploaded Python 3

File details

Details for the file benchmarkingcli-0.4.0.tar.gz.

File metadata

  • Download URL: benchmarkingcli-0.4.0.tar.gz
  • Upload date:
  • Size: 2.1 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for benchmarkingcli-0.4.0.tar.gz
Algorithm Hash digest
SHA256 d395ba83a22e9de64d8aa80943193896e82b0ed37b7a9194b0e82f0c18fd2b27
MD5 a074da98e46dba00069f2934a0ac2898
BLAKE2b-256 ba4ff69d58b1b6243922f8f8451dee62ed6e05c7c461553adf65f6f1cb95c247

See more details on using hashes here.

File details

Details for the file benchmarkingcli-0.4.0-py3-none-any.whl.

File metadata

File hashes

Hashes for benchmarkingcli-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 69be3bcea07c0439ac244ae94048fd7bc179e13df7332d83749f4e36ed7da9cc
MD5 a8ceda45022df48789d87a42c75b64d3
BLAKE2b-256 8ca3eb9ce79c3a728641e5029ba2fb966c9364e3ec025eba6efb97ec17bde501

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page