Data Overload Vol. 15

Welcome to Data Overload Vol. 15, Orchest’s monthly newsletter.

We improved the stability of Orchest Cloud 🪨

To improve the stability of Orchest Cloud for instances with many batch jobs and pipelines, we have made some infrastructure changes to make sure the CPU resource requests of containers are honored properly. Let us know your thoughts in our public Slack community!

See you at PyData NYC! 🗽

Next week PyData NYC, one of the biggest and oldest PyData events, will take place at the Microsoft Conference Center, next to Times Square. Our Data Scientist Advocate will cross the pond to speak about Expressive and fast dataframes in Python with Polars. With four to five tracks packed with amazing talks, it will surely be a blast. See you there!

Working on a new file picker 🏗️

Our Frontend Team has been working hard on a new file picker, featuring a fresh design, great keyboard support, and the capability of creating new files directly from it. Stay tuned!

Thank you!

Our community keeps growing and we keep learning about our users use cases, their desires for new features, and how much they love the product. Thanks a lot!

• Juan Luis Cano (Data Scientist Advocate)

Product updates

🎥 New demo video

We recorded a new Orchest video updated with the new UI. Check it out in our homepage!

🛣️ Feature roadmap

Curious about what’s next in Orchest? Head to our feature roadmap!

What we're reading

• Scaling Kubernetes to 2,500 Nodes (blog post)
  
  We always enjoy learning more about how big engineering teams troubleshoot and improve their infrastructure. This excellent writeup covers optimizing the latency of etcd, improving the KubeDNS configuration, optimizing Docker pulls, and more. Highly recommended!