scitex-resource 0.3.3

System resource info, processor usage logging, RAM limit — standalone module from the SciTeX ecosystem

scitex-resource

System resource info, processor usage logging, and RAM limiting helpers, extracted from the SciTeX ecosystem as a standalone package.

Install

pip install scitex-resource

API

import scitex_resource as r

# Flat hub-friendly metrics (cpu/mem/disk/gpu/load) — cross-platform via psutil
metrics = r.get_metrics()

# Canonical machine identity — every scitex-* package consumes this so they
# all agree on "what host am I". Resolves env > project config > user config
# > short hostname. See ~/.scitex/resource/config.yaml below.
name = r.get_machine_name()
cfg = r.get_machine_config()        # {"canonical_name", "aliases", "role", "hpc": {...}}

# Rich human-readable snapshot (system info, GPU, network, disk partitions)
specs = r.get_specs()

# CPU/RAM samples + continuous CSV logging
usage = r.get_processor_usages()
r.log_processor_usages("/tmp/usage.csv", limit_min=30, interval_s=1)

# Cap process RAM
r.limit_ram(0.5)

Machine identity config — ~/.scitex/resource/config.yaml

machine:
  canonical_name: mba                  # what every scitex-* package uses to refer to this host
  aliases:                              # optional; cross-package discovery / drift detection
    - Yusukes-MacBook-Air
    - Yusukes-MacBook-Air.local
  role: head                            # generic role tag (head, worker, hpc-login, ...)
  hpc:                                  # optional; HPC-only
    cluster: spartan
    login_only: true                    # don't surface login-node CPU as available
    partitions: [physical, sapphire]

Resolution cascade (highest precedence first):

1. $SCITEX_RESOURCE_MACHINE
2. <project>/.scitex/resource/config.yaml machine.canonical_name
3. ~/.scitex/resource/config.yaml machine.canonical_name
4. socket.gethostname().split(".", 1)[0]

This is the ecosystem convention — see scitex-python _skills/general/01_arch_06_local-state-directories.md for the full .scitex/<pkg-short>/ layout (config tracked, runtime/ ignored).

Status

Standalone fork of scitex.resource. Deps: pandas, psutil, PyYAML, matplotlib.

Decoupling notes:

• scitex.str.readable_bytes / scitex.gen.fmt_size / scitex.str.printc → vendored as 3 small helpers in _compat.py.
• scitex.io._load.load / scitex.io._save.save → use pandas.read_csv / to_csv directly for the CSV log files; defer to scitex.io only if a non-CSV path is requested (raises ImportError without scitex installed).
• scitex.sh.sh → prefer scitex_sh if installed, fall back to plain subprocess.run (list-only).
• scitex.session.start/close __main__ block → simplified to plain main().

The umbrella package's scitex.resource import path is preserved via a sys.modules-alias bridge. 65/65 tests pass.

License

AGPL-3.0-only (see LICENSE).