Skip to main content

Performance monitoring and system utilities

Project description

warpt

A unified command-line tool for hardware discovery, stress testing, and performance monitoring.

warpt provides a vendor-agnostic interface for understanding and validating computational resources—answering questions like "What hardware do I have?", "Is it working correctly?", and "How fast is it?"

Installation

pip install warpt

For stress testing capabilities:

pip install warpt[stress]

Requirements: Python 3.8+ (3.11+ recommended) | Linux, macOS, or Windows

Quick Start

# Discover your hardware
warpt list

# Run CPU stress tests
warpt stress -c cpu

# Monitor system in real-time
warpt monitor

# Check power consumption (Linux/macOS)
warpt power

Features

Command Description
warpt list Detect CPU, GPU, memory, storage, and installed ML frameworks
warpt stress Run stress tests across CPU, GPU, RAM, storage, and network
warpt monitor Real-time system monitoring with TUI dashboard
warpt power Power consumption monitoring and per-process attribution
warpt carbon Track energy consumption, CO2 emissions, and estimated cost
warpt benchmark Performance benchmarking suite

Documentation

Platform Support

Platform Status
Linux Full support
macOS Full support (power monitoring requires sudo)
Windows Limited support (see Known Limitations)

GPU Support: NVIDIA GPUs supported. AMD, Intel, and Apple Silicon GPU support coming soon.

Carbon Tracking

warpt automatically tracks energy usage and CO2 emissions during stress tests and power monitoring. You can also track any workload manually:

# Automatic — built into stress tests
warpt stress -c cpu -d 30
# [carbon] 30.2s | 23.8W avg | 199.7 mWh | 0.08g CO2 | $0.0000 | less than breathing for a minute

# Manual — track any workload
warpt carbon start
# ... run your workload ...
warpt carbon stop

# View history and totals
warpt carbon history
warpt carbon summary

# Check available grid regions and carbon intensities
warpt carbon regions

Carbon calculations use regional grid intensity data to estimate CO2 emissions from energy consumption. Configure your region with --region (defaults to US).

Example Output

$ warpt list

CPU Information:
  Make:               Intel
  Model:              Xeon W-2295
  Architecture:       x86_64

Topology:
  Total Sockets:      1
  Total Phys Cores:   18
  Total Logic Cores:  36

Memory Information:
  Total:              128.0 GB
  Type:               DDR4

GPU Information:
  GPU 0: NVIDIA RTX 4090
    Memory: 24576 MB
    Driver: 545.23.08

Alpha Release

This is an alpha release. Some features are still in development:

  • Carbon tracking — new in v0.2.0
  • AMD GPU support (ROCm) — in progress
  • Intel GPU support (oneAPI) — in progress
  • Apple Neural Engine — in progress
  • Additional benchmarks — expanding

See the Support Matrix for full details.

Feedback

We'd love to hear from you:

License

MIT License — see LICENSE for details.

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

warpt-0.2.0.tar.gz (190.8 kB view details)

Uploaded Source

Built Distribution

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

warpt-0.2.0-py3-none-any.whl (248.7 kB view details)

Uploaded Python 3

File details

Details for the file warpt-0.2.0.tar.gz.

File metadata

  • Download URL: warpt-0.2.0.tar.gz
  • Upload date:
  • Size: 190.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.0

File hashes

Hashes for warpt-0.2.0.tar.gz
Algorithm Hash digest
SHA256 77a10a9d244f1e3b5a1a4d4214573b991d7e41f76d5b58972681a2ac62aadf3c
MD5 7c4f8a05ff689aba35e105c152b0f30d
BLAKE2b-256 bbcd133ca74296fc07fd57b2f3fbe9dca22b82ce2e3dd0b85b847b8158167309

See more details on using hashes here.

File details

Details for the file warpt-0.2.0-py3-none-any.whl.

File metadata

  • Download URL: warpt-0.2.0-py3-none-any.whl
  • Upload date:
  • Size: 248.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.0

File hashes

Hashes for warpt-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 3845578d072b5fb5edc1b2fb591c52026d9e1c51fa06d2166fdd5fa11aabfe60
MD5 f69786108d4f05a738adfe3d0fe12ae2
BLAKE2b-256 8ba1b02736809dd6fc21ef6727b2a42d659f2277fc00822726f786e510252aa0

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