<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w" -->

---
title: Netflix Adopts Cloud-Native Job Queueing System Kueue to...
description: Netflix migrated most of its batch workloads from its homegrown Compute Managed Batch (CMB) system, built on Titus in 2018, to Kueue, an open-source...
canonical: https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Netflix Adopts Cloud-Native Job Queueing System Kueue to Replace an In-House Solution | daily.dev
og:description: Netflix migrated most of its batch workloads from its homegrown Compute Managed Batch (CMB) system, built on Titus in 2018, to Kueue, an open-source...
og:url: https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w
og:image: https://api.daily.dev/og/posts/7GBkJcG1w.png
og:image:alt: Netflix Adopts Cloud-Native Job Queueing System Kueue to Replace an In-House Solution
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.

# Netflix Adopts Cloud-Native Job Queueing System Kueue to Replace an In-House Solution

**[InfoQ](https://daily.dev/sources/infoq)** · 3 min read · 0 upvotes · 0 comments

## Summary

Netflix migrated most of its batch workloads from its homegrown Compute Managed Batch (CMB) system, built on Titus in 2018, to Kueue, an open-source Kubernetes-native job queueing system. Engineers mapped CMB's tenant hierarchies to Kueue's Cohorts, ClusterQueue, and LocalQueue resources, maintained API parity to keep the migration transparent to users, and migrated the largest and most complex customer first to build confidence, completing that production migration in 4 weeks. Netflix now manages millions of batch workloads on Kueue and has used preemption-based fair sharing to improve average resource utilization, though the migration is still ongoing.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.infoq.com/news/2026/08/netflix-kueue-kubernetes-batch>

## Questions this post answers

### Why did Netflix replace its in-house Compute Managed Batch system with Kueue?

Netflix replaced Compute Managed Batch (CMB), built in 2018 on its Titus container platform, with Kueue because many of CMB's key features had since been absorbed into open-source Kubernetes projects, and extending CMB had become harder since it wasn't tightly integrated with Kubernetes. Kueue offered wider adoption, more extensive capabilities, and faster innovation.

_Teams weighing homegrown infrastructure against open-source alternatives can track real migration case studies like this on daily.dev._

### How did Netflix map its CMB tenant model onto Kueue's resource model?

Netflix mapped internal CMB tenants to Kueue Cohorts and individual leaf tenants to a paired ClusterQueue and LocalQueue resource in Kueue. Resource flavors and nominal quotas were then used to reproduce the capacity requirements that had previously been configured in CMB, allowing a like-for-like transition of quota and scheduling behavior.

_Engineers planning a similar Kubernetes batch scheduling migration can find architecture breakdowns like this via daily.dev._

### What lesson did Netflix learn about sequencing a large-scale system migration?

Migrate the most complex use case first rather than saving it for last. Netflix deliberately moved its largest and most complicated customer onto Kueue early, which built confidence for migrating the rest and let the entire production migration finish in just four weeks, alongside maintaining strict API parity to reduce risk.

_Anyone sequencing a risky infrastructure migration can study real-world lessons like this through daily.dev._

## Similar posts on daily.dev

- [AI21 trains its models on AI Hypercomputer](https://daily.dev/posts/ai21-trains-its-models-on-ai-hypercomputer-skzkrwjlk) · Google Cloud · 0 upvotes · 0 comments
- [Manage LLM evaluation workloads at scale with EvalHub and Kueue](https://daily.dev/posts/manage-llm-evaluation-workloads-at-scale-with-evalhub-and-kueue-dpquikccd) · Red Hat Developer · 1 upvotes · 0 comments
- [Red Hat build of Kueue 1.3: Enhanced batch workload management on Kubernetes](https://daily.dev/posts/red-hat-build-of-kueue-1-3-enhanced-batch-workload-management-on-kubernetes-xslvtp3us) · Red Hat Developer · 0 upvotes · 0 comments
- [GPU Job Queueing with Kueue and DRA: Scheduling AI Workloads Without Idle Capacity](https://daily.dev/posts/gpu-job-queueing-with-kueue-and-dra-scheduling-ai-workloads-without-idle-capacity-g1csd6qkw) · Cast AI · 0 upvotes · 0 comments
- [Netflix Moves Toward Open Source Flink Autoscaler for 30,000\+ Streaming Jobs](https://daily.dev/posts/netflix-moves-toward-open-source-flink-autoscaler-for-30-000-streaming-jobs-u4oedc6lc) · InfoQ · 0 upvotes · 0 comments

---

Tags: [#kubernetes](https://daily.dev/tags/kubernetes), [#cloud-native](https://daily.dev/tags/cloud-native), [#netflix](https://daily.dev/tags/netflix)

[View this post on daily.dev](https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w)

```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":"Netflix Adopts Cloud-Native Job Queueing System Kueue to Replace an In-House Solution","url":"https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w"},"datePublished":"2026-08-12T14:42:45.660Z","dateModified":"2026-09-14T07:41:54.987Z","description":"Netflix migrated most of its batch workloads from its homegrown Compute Managed Batch (CMB) system, built on Titus in 2018, to Kueue, an open-source...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5387bfb91e3af9cfde39bdf8c7a24a26?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5387bfb91e3af9cfde39bdf8c7a24a26?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"InfoQ","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":"InfoQ","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/afc3bced3e1e4b188dd9127017a60e0c","url":"https://daily.dev/sources/infoq"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"kubernetes,cloud-native,netflix","timeRequired":"PT3M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"InfoQ","item":"https://daily.dev/sources/infoq"},{"@type":"ListItem","position":3,"name":"Netflix Adopts Cloud-Native Job Queueing System Kueue to Replace an In-House Solution"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/netflix-adopts-cloud-native-job-queueing-system-kueue-to-replace-an-in-house-solution-7gbkjcg1w#faq","mainEntity":[{"@type":"Question","name":"Why did Netflix replace its in-house Compute Managed Batch system with Kueue?","acceptedAnswer":{"@type":"Answer","text":"Netflix replaced Compute Managed Batch (CMB), built in 2018 on its Titus container platform, with Kueue because many of CMB's key features had since been absorbed into open-source Kubernetes projects, and extending CMB had become harder since it wasn't tightly integrated with Kubernetes. Kueue offered wider adoption, more extensive capabilities, and faster innovation. Teams weighing homegrown infrastructure against open-source alternatives can track real migration case studies like this on daily.dev."}},{"@type":"Question","name":"How did Netflix map its CMB tenant model onto Kueue's resource model?","acceptedAnswer":{"@type":"Answer","text":"Netflix mapped internal CMB tenants to Kueue Cohorts and individual leaf tenants to a paired ClusterQueue and LocalQueue resource in Kueue. Resource flavors and nominal quotas were then used to reproduce the capacity requirements that had previously been configured in CMB, allowing a like-for-like transition of quota and scheduling behavior. Engineers planning a similar Kubernetes batch scheduling migration can find architecture breakdowns like this via daily.dev."}},{"@type":"Question","name":"What lesson did Netflix learn about sequencing a large-scale system migration?","acceptedAnswer":{"@type":"Answer","text":"Migrate the most complex use case first rather than saving it for last. Netflix deliberately moved its largest and most complicated customer onto Kueue early, which built confidence for migrating the rest and let the entire production migration finish in just four weeks, alongside maintaining strict API parity to reduce risk. Anyone sequencing a risky infrastructure migration can study real-world lessons like this through daily.dev."}}]}
```

