# Ray distributed computing framework

10 Hacker News commenters recommended Ray distributed computing framework and 7 criticized it, November 2025 to November 2026. It was recommended in 6 of 13 months.

**Verdict:** People recommend Ray for scaling Python workloads across machines, especially ML infrastructure, with built-in autoscaling. Critics say clusters add heavy overhead and cost, and that it is slow.

## Why people recommend it

- Strong fit for ML infrastructure (4 comments): "Ray is now becoming de-facto framework for compositional learning infrastructure" ([source](https://news.ycombinator.com/item?id=46123573)) "have also noticed this pattern in Modal [1] and Ray [2] and all for it." ([source](https://news.ycombinator.com/item?id=46859990)) "there are libraries and clear documentation (e.g. Lightning, Ray, etc)" ([source](https://news.ycombinator.com/item?id=46285465))
- Built-in autoscaling for clusters (2 comments): "stitched together with Ray autoscaling nodes it can get really nice" ([source](https://news.ycombinator.com/item?id=48893702)) "Ray has autoscaling, so it's very good for this kind of thing." ([source](https://news.ycombinator.com/item?id=45781564))

## What people warn about

- Cluster overhead and high costs (2 comments): "managing the infrastructure overhead of Celery or Ray is a massive time sink" ([source](https://news.ycombinator.com/item?id=47376168)) "ray clusters don't scale well and end up costing you more money" ([source](https://news.ycombinator.com/item?id=46826532))
- Too slow compared with alternatives (2 comments): "it's faster than prefect and ray so I can use it for more stuff without having to import a 10000000 line library" ([source](https://news.ycombinator.com/item?id=47445032)) "Dask and Ray were also way too slow" ([source](https://news.ycombinator.com/item?id=45926420))

## Alternatives hackers recommend

- [Apache Spark](https://hackerapproved.com/p/apache-spark.md): 18 recommend
- [Apache Hadoop](https://hackerapproved.com/p/hadoop.md): 8 recommend

## Cite as

Hacker Approved, "Ray distributed computing framework", https://hackerapproved.com/p/ray/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).