{"paper":{"title":"CudaMon: An R Package to Monitor NVIDIA GPUs, Showcased by Monitoring a GPU-accelerated Single-cell Analysis Workflow in R","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CudaMon is an R package that monitors NVIDIA GPU metrics in real time for accelerated computational workflows.","cross_cats":[],"primary_cat":"stat.CO","authors_text":"Davide Risso, Gabriele Sales, Mohammad Amin Zadenoori, Riccardo Ceccaroni","submitted_at":"2026-05-13T15:47:21Z","abstract_excerpt":"NVIDIA GPUs have recently started to be used in computational biology, yet R users lack integrated GPU monitoring tools, forcing reliance on external utilities like nvidia-smi. We introduce CudaMon, an R package providing real-time monitoring of GPU utilization, memory, temperature, and power draw via NVML, along with data export and visualization utilities. Monitoring a GPU-accelerated single-cell RNA-seq pipeline (1M brain cells, RAPIDS workflow) shows that compute-intensive steps (PCA, UMAP, t-SNE) exceed 90% GPU utilization, while data management phases reveal bottlenecks. CudaMon facilita"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We introduce CudaMon, an R package providing real-time monitoring of GPU utilization, memory, temperature, and power draw via NVML, along with data export and visualization utilities. Monitoring a GPU-accelerated single-cell RNA-seq pipeline (1M brain cells, RAPIDS workflow) shows that compute-intensive steps (PCA, UMAP, t-SNE) exceed 90% GPU utilization, while data management phases reveal bottlenecks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The NVML-based monitoring accurately captures GPU metrics in real time without adding meaningful overhead or interfering with the monitored GPU-accelerated computations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CudaMon is a new R package enabling direct real-time GPU monitoring for R users to optimize and debug GPU-accelerated workflows, demonstrated on a large single-cell RNA-seq pipeline.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CudaMon is an R package that monitors NVIDIA GPU metrics in real time for accelerated computational workflows.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1525b74372251fd986cdb0724921d40c6d956862202c0fa2c79534b9bb12523a"},"source":{"id":"2605.13928","kind":"arxiv","version":1},"verdict":{"id":"32585079-ab0f-4eca-956a-9cebd70796d5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:33:14.692608Z","strongest_claim":"We introduce CudaMon, an R package providing real-time monitoring of GPU utilization, memory, temperature, and power draw via NVML, along with data export and visualization utilities. Monitoring a GPU-accelerated single-cell RNA-seq pipeline (1M brain cells, RAPIDS workflow) shows that compute-intensive steps (PCA, UMAP, t-SNE) exceed 90% GPU utilization, while data management phases reveal bottlenecks.","one_line_summary":"CudaMon is a new R package enabling direct real-time GPU monitoring for R users to optimize and debug GPU-accelerated workflows, demonstrated on a large single-cell RNA-seq pipeline.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The NVML-based monitoring accurately captures GPU metrics in real time without adding meaningful overhead or interfering with the monitored GPU-accelerated computations.","pith_extraction_headline":"CudaMon is an R package that monitors NVIDIA GPU metrics in real time for accelerated computational workflows."},"references":{"count":15,"sample":[{"doi":"","year":2020,"title":"I tried a bunch of things: the dangers of unexpected overfitting in classification , author=. bioRxiv , volume=. 2020 , publisher=","work_id":"8b1421cc-c552-4caa-a348-4279eefe894b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2012,"title":"Communications of the ACM , volume=","work_id":"63b972b6-7dc6-4413-8971-b07fe3c427d9","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Mass Spectrometry Data Analysis in Proteomics , pages=","work_id":"1f954926-7a02-4bbf-b886-dbe2ae61c450","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Cazzaniga, Paolo and Besozzi, Daniela and Merelli, Ivan and Manzoni, Luca , volume=. 2020 , publisher=","work_id":"724ef8c4-ab1d-4d89-8e9e-147f4a79c389","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2014,"title":"Scientific Visualization , author=. 2014 , publisher=","work_id":"958ddffe-1b94-403c-b661-8e6e0f232e0e","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":15,"snapshot_sha256":"0e57a426f814b01f2ffc309bb84254af83f88c526752413a342e5c79f68b83e0","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}