--- title: How Much of a Data Science Workflow Can Run on a GPU... description: GPU acceleration is now practical for tabular data workflows in Python, not just deep learning. Using the NYC Yellow Taxi dataset (~9-10M rows), this post... canonical: https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How Much of a Data Science Workflow Can Run on a GPU Today? Part 1: Accelerating Data Preparation | daily.dev og:description: GPU acceleration is now practical for tabular data workflows in Python, not just deep learning. Using the NYC Yellow Taxi dataset (~9-10M rows), this post... og:url: https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt og:image: https://api.daily.dev/og/posts/0T7hiJMlT.png og:image:alt: How Much of a Data Science Workflow Can Run on a GPU Today? Part 1: Accelerating Data Preparation og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # How Much of a Data Science Workflow Can Run on a GPU Today? Part 1: Accelerating Data Preparation **[Towards Data Science](https://daily.dev/sources/tds)** · 10 min read · 0 upvotes · 0 comments ## Summary GPU acceleration is now practical for tabular data workflows in Python, not just deep learning. Using the NYC Yellow Taxi dataset (~9-10M rows), this post walks through three approaches: the native cuDF API (pandas-like GPU DataFrames), cudf.pandas (zero-code-change accelerator for existing pandas code), and the Polars GPU engine (pass engine='gpu' to collect()). All three support automatic CPU fallback for unsupported operations. Benchmarks show significant speedups for groupby aggregations, joins, and sorting on large datasets, while CPUs remain competitive for smaller data due to GPU transfer overhead. The Polars GPU engine also supports multi-GPU scaling via RayEngine. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Using Accelerated Computing to Live-Steer Scientific Experiments at Massive Research Facilities](https://daily.dev/posts/using-accelerated-computing-to-live-steer-scientific-experiments-at-massive-research-facilities-hneuhgb21) · NVIDIA Developer · 0 upvotes · 0 comments - [Crunching data? Put that laptop hardware to use](https://daily.dev/posts/crunching-data-put-that-laptop-hardware-to-use-1fcnht8lv) · Simple Thread · 0 upvotes · 0 comments - [A new streaming backend for the Polars GPU engine](https://daily.dev/posts/a-new-streaming-backend-for-the-polars-gpu-engine-m46jtgep2) · Polars Blog · 0 upvotes · 0 comments - [What Can We Do When Memory Becomes the New Bottleneck in Data Engineering?](https://daily.dev/posts/what-can-we-do-when-memory-becomes-the-new-bottleneck-in-data-engineering--h1dho6aid) · Towards Data Science · 0 upvotes · 0 comments --- Tags: [#data-science](https://daily.dev/tags/data-science), [#gpu](https://daily.dev/tags/gpu), [#pandas](https://daily.dev/tags/pandas), [#polars](https://daily.dev/tags/polars) [View this post on daily.dev](https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"How Much of a Data Science Workflow Can Run on a GPU Today? Part 1: Accelerating Data Preparation","url":"https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt"},"datePublished":"2026-07-21T13:34:20.676Z","dateModified":"2026-07-21T13:34:42.993Z","description":"GPU acceleration is now practical for tabular data workflows in Python, not just deep learning. Using the NYC Yellow Taxi dataset (~9-10M rows), this post...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/136e1bdf88c42caa8459000fe142cc4d?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/136e1bdf88c42caa8459000fe142cc4d?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Towards Data Science","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Towards Data Science","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tds","url":"https://daily.dev/sources/tds"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-much-of-a-data-science-workflow-can-run-on-a-gpu-today-part-1-accelerating-data-preparation-0t7hijmlt","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"data-science,gpu,pandas,polars","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Towards Data Science","item":"https://daily.dev/sources/tds"},{"@type":"ListItem","position":3,"name":"How Much of a Data Science Workflow Can Run on a GPU Today? Part 1: Accelerating Data Preparation"}]} ```