Skip to content

Changelog

v0.5.0 (April 10, 2026)#

  • Add optional and configurable metric streaming to the Spare Cores Sentinel: pass sentinel_token to ResourceTracker (or set SENTINEL_API_TOKEN env var) to periodically upload gzipped CSV batches to the Sentinel object storage, with automatic STS credential refresh and inline CSV fallback for short runs.
  • Add sentinel_api module: HTTP client for registering runs, refreshing credentials, and submitting final data.
  • Add s3_upload module: lightweight S3 uploader using temporary STS credentials.
  • Fix macOS disk space reporting: correctly use the APFS data volume (/System/Volumes/Data) instead of summing all partitions; switch disk space calculation from binary (1024^3) to SI (10^9).
  • Fix memory column naming and units: rename memory → memory_mib (process), memory_free/memory_used/etc. → *_mib suffix (system), gpu_vram → gpu_vram_mib; all memory metrics now in MiB.
  • Fix process I/O column naming: rename read_bytes/write_bytes → disk_read_bytes/disk_write_bytes.
  • Add offset parameter to get_combined_metrics for incremental reads.
  • Add context manager support to ResourceTracker.

!! Breaking changes:

  • The autostart parameter of ResourceTracker now defaults to False, but it's still auto-started when used in Metaflow or as a context manager.

v0.4.2 (August 8, 2025)#

  • Add cleanup method to ResourceTracker to cleanup temp files and background processes.

v0.4.1 (August 8, 2025)#

Introduce native R support. Example usage:

library(resource_tracker)

tracker <- ResourceTracker$new()
tracker$wait_for_samples(1)
tracker$stats()
tracker$recommend_resources()
tracker$recommend_server()
tracker$report()$browse()

Find more details in the R integration docs.

Additional changes:

  • Split documentation and extend into multiple pages with more details on the integrations.

v0.4.0 (August 6, 2025)#

Refactoring and cleanup release with the main focus on extracting the reporting features from the Metaflow extension and support it in standalone use as well.

Main changes:

  • Much better support for standalone use of ResourceTracker via the stats, recommend_resources, recommend_server and report methods

More details:

  • Rename PidTracker to ProcessTracker to better reflect its purpose. PidTracker is still available as an alias that is to be deprecated in the future.
  • Rename get_pid_stats to get_process_stats and related references in ResourceTracker and ProcessTracker.
  • Rename the pid_tracker and system_tracker properties of ResourceTracker to process_metrics and system_metrics respectively. All related references were also updated, e.g. in the Metaflow extension and docs.
  • Rename process-related helpers in the ProcFS implementation from pid prefix to process prefix.
  • Fix SystemTracker and ProcessTracker to not print dummy stats on start when header is disabled
  • Add optional start_time parameter to SystemTracker and ProcessTracker
  • Update ResourceTracker to start tracking at the nearest interval in the future, syncing SystemTracker and ProcessTracker
  • Fix SystemTracker and ProcessTracker to not drift by a few nanoseconds in every interval
  • Move cloud and server discovery along with the server allocation check to the ResourceTracker class from the Metaflow-specific decorators
  • Extract serialization and deserialization of ResourceTracker from the Metaflow extension into the ResourceTracker class with snapshot and dump(s)/load(s) methods
  • Round timestamp and user/system time to reasonable (6/4) decimal places
  • Rework internal data structure of TinyDataFrame to use a list of lists instead of a dictionary of lists to support more efficient slicing and column renaming
  • Add get_combined_metrics method to ResourceTracker to combine process_metrics and system_metrics into a single data frame, optionally with all metrics converted to bytes, and columns renamed to use human-friendly names
  • Add stats method to TinyDataFrame to compute on-demand statistics on columns
  • Add stats method to ResourceTracker to compute statistics on the combined metrics
  • Add minimal support for Handlebars-like templates in the render_template function
  • Add report method to ResourceTracker to generate a report in HTML format, and use that from the Metaflow extension
  • Add recommend_resources and recommend_server methods to ResourceTracker
  • Windows-specific reliability improvements

Related breaking changes:

  • Historical data collected before v0.4.0 is not compatible with the new ResourceTracker class, and will be discarded
  • TinyDataFrame is no longer made available from resource_tracker directly, but from the resource_tracker.tiny_data_frame submodule

v0.3.1 (May 30, 2025)#

  • Generate card for failed step
  • Note failed step status in card
  • Standardize timestamp format and timezone

v0.3.0 (March 27, 2025)#

  • Extract background process management and related complexities from the track_resources decorator into the ResourceTracker class to track resource usage of a process and/or the system in a non-blocking way
  • Add unit tests for the ResourceTracker class, including checks for deadlocks and partially started trackers
  • Keep test HTML card as GHA artifacts for manual inspection
  • Improve documentation

v0.2.1 (March 21, 2025)#

  • Fix don't always round up CPU/GPU recommendations
  • Improve error message on missing historical data
  • Improve documentation

v0.2.0 (March 21, 2025)#

Relatively major package rewrite to support alternative tracker implementations (other than directly reading from /proc). No breaking changes in the public API on Linux.

  • Add tracker implementation using psutil to support MacOS and Windows
  • Fix data issues with the /proc implementation after validating with the psutil version (e.g. number of processes reported)
  • Refactor code for better maintainability
  • Add additional unit tests:
    • Tracker implementation using procfs
    • Tracker implementation using psutil
    • Consistency between tracker implementations
    • Metaflow decorators
  • Extend CI/CD pipeline:
    • Test on Linux, MacOS, and Windows
    • Test multiple Python versions (3.9, 3.10, 3.11, 3.12, 3.13)
  • Improve documentation

v0.1.2 (March 18, 2025)#

  • Add experimental psutil support
  • Add server info card for operating system

v0.1.1 (March 17, 2025)#

  • Fix rounding down recommended vCPUs with <0.5 load
  • Add info popups with more details and disclaimers for recommendations
  • Add detection for shared server environments
  • Add potential cost savings card
  • Improve documentation

v0.1.0 (March 12, 2025)#

Initial PyPI release of resource-tracker with the following features:

  • Detect if the system is running on a cloud provider, and if so, detect the provider, region, and instance type
  • Detect main server hardware (CPU count, memory amount, disk space, GPU count and VRAM amount)
  • Track system-wide resource usage:
    • Process count
    • CPU usage (user + system time, relative vCPU percentage)
    • Memory usage (total, free, used, buffers, cached, active anon, inactive anon pages)
    • Disk I/O (read and write bytes)
    • Disk space usage (total, used, free)
    • Network I/O (receive and transmit bytes)
    • GPU and VRAM usage (using nvidia-smi)
  • Track resource usage of a process and its descendant processes:
    • Descendant process count
    • CPU usage (user + system time, relative vCPU percentage)
    • Memory usage (based on proportional set sizes)
    • Disk I/O (read and write bytes)
    • GPU and VRAM usage (using nvidia-smi pmon)
  • Add Metaflow plugin for tracking resource usage of a step:
    • Track process and system resource usage for the duration of the step
    • Generate a card with the resource usage data
    • Suggest @resources decorator for future runs
    • Find cheapest cloud instance type for a step