We help small teams achieve big things.
We are a UK based, fully remote consultancy who specialise in Data, Analytics & AI, and Cloud Native App Dev on Microsoft Fabric & Azure, and are a .NET Foundation Corporate Sponsor.
We produce two free weekly newsletters;
☁️ Azure Weekly - azureweekly.info for all things about the Microsoft Azure Platform,
📈Power BI Weekly - powerbiweekly.info for all things data visualisation and Power Platform.
Keep up with everything that's going on at endjin via our blog:
👉 endjin.com/blog
👉 endjin.com/talks
👉 twitter.com/endjin
👉 www.linkedin.com/company/endjin
Information about our Open Source projects can be found at endjin.com/open-source
Find out more at endjin.com
#Microsoft #MicrosoftFabric #PowerBI #AI #Data #Analytics #DevOps #Azure #Cloud
endjin
Barry Smart analyses the results of benchmarking different data processing engines on #MicrosoftFabric.
We compare #Pandas, #PySpark, #Polars, and #DuckDB across various compute configurations.
The results provide concrete, Fabric-specific evidence for a broader industry trend: for medium-scale datasets (anything up to ~100GB), modern in-process engines like DuckDB and Polars on single-node Python notebooks are consistently faster and up to 5x cheaper than distributed Spark clusters.
The code used to generate the benchmark is available in a public repo on GitHub.
endjin.com/blog/fabric-performance-benchmarking-sp…
1 week ago (edited) | [YT] | 1
View 0 replies
endjin
We've just published a 4-part series about our experience migrating from PySpark / Pandas to Polars and how we got a huge performance improvement, lower costs, and happier endjin-eers!
endjin.com/blog/2026/01/polars-faster-pipelines-si…
2 months ago | [YT] | 1
View 0 replies
endjin
A deep dive into the new Implicit Span Conversion feature in C# 14 - this contains important information for anyone who maintains .NET libraries.
4 months ago | [YT] | 1
View 0 replies