--- title: Cluster Management: keep the controls, skip the cluster ops description: Tinybird introduces Cluster Management, a feature for Dedicated infrastructure Organizations that gives teams direct control over ClickHouse replica capacity... canonical: https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Cluster Management: keep the controls, skip the cluster ops | daily.dev og:description: Tinybird introduces Cluster Management, a feature for Dedicated infrastructure Organizations that gives teams direct control over ClickHouse replica capacity... og:url: https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a og:image: https://api.daily.dev/og/posts/5CTiRXu5a.png og:image:alt: Cluster Management: keep the controls, skip the cluster ops 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. # Cluster Management: keep the controls, skip the cluster ops **[Tinybird](https://daily.dev/sources/tinybird)** · 7 min read · 2 upvotes · 0 comments ## Summary Tinybird introduces Cluster Management, a feature for Dedicated infrastructure Organizations that gives teams direct control over ClickHouse replica capacity and workload distribution while Tinybird continues to operate the underlying platform. Teams can observe hardware metrics via Service Data Sources, then add, resize, or remove replicas and rebalance read/write/copyjob weights through the Organization UI or the Organizations API. Resizing follows an add-reroute-verify-remove workflow that avoids downtime, with old replicas serving traffic until new ones are verified. Weight-based routing uses proportional round-robin, and API mutations require fresh old_weights to prevent conflicting concurrent changes. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do read and write weights distribute traffic across ClickHouse replicas in Tinybird's Cluster Management? Weights use proportional, weighted round-robin distribution, so a read weight of 2 receives twice the query traffic of a read weight of 1. Reader weights control query traffic, writer weights control Events API and Kafka ingestion, and copyjob weights control Copy Pipe jobs; each weight ranges from 0 to 65535, and at least one replica must have a non-zero read weight and one a non-zero write weight. _Teams tuning ClickHouse replica traffic can find this kind of operational detail curated on daily.dev._ ### How can I resize a ClickHouse cluster on Tinybird without downtime? Use an add, reroute, verify, remove workflow: add replicas at the target CPU and memory size, wait for them to become available, rebalance read and write traffic onto them, verify query latency and ingestion throughput, then remove the old replicas. Old replicas keep serving until removed, so the cluster stays available, though both old and new capacity are billed during the overlap, which typically lasts a few minutes. _Engineers planning zero-downtime database resizing can track workflows like this on daily.dev._ ### How do I query CPU usage metrics for a Tinybird Dedicated ClickHouse cluster? Query the organization.metrics_logs Service Data Source, which receives new records every minute and is free to query. An example selects cluster, host, and average CPU usage percentage over the last 5 minutes grouped by cluster and host; the same source also exposes memory, executing queries, system load, and replication queue metrics, with organization.endpoint_errors covering API errors. _Developers monitoring database cluster health can follow practical observability approaches like this on daily.dev._ ## Similar posts on daily.dev - [Self-Serve Replicas for Dedicated Infrastructure: Now in Private Beta](https://daily.dev/posts/self-serve-replicas-for-dedicated-infrastructure-now-in-private-beta-jjqe46uga) · Tinybird · 3 upvotes · 0 comments - [How Tinybird's storage architecture works: S3, local caching, and zero-copy replication](https://daily.dev/posts/how-tinybird-s-storage-architecture-works-s3-local-caching-and-zero-copy-replication-uojcvceuc) · Tinybird · 3 upvotes · 0 comments - [How We Built Branches: Sharing Production Data Without Copying It](https://daily.dev/posts/how-we-built-branches-sharing-production-data-without-copying-it-2mlvlk1ee) · Tinybird · 1 upvotes · 0 comments - [Building more resilient ingestion with smart backpressure handling](https://daily.dev/posts/building-more-resilient-ingestion-with-smart-backpressure-handling-2ayxahm6m) · Tinybird · 0 upvotes · 0 comments --- Tags: [#architecture](https://daily.dev/tags/architecture), [#observability](https://daily.dev/tags/observability), [#clickhouse](https://daily.dev/tags/clickhouse), [#tinybird](https://daily.dev/tags/tinybird) [View this post on daily.dev](https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a) ```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":"Cluster Management: keep the controls, skip the cluster ops","url":"https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a"},"datePublished":"2026-09-03T13:05:57.273Z","dateModified":"2026-09-13T20:36:38.499Z","description":"Tinybird introduces Cluster Management, a feature for Dedicated infrastructure Organizations that gives teams direct control over ClickHouse replica capacity...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/86666e836aa32d5a13e1e2e9220507ac?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/86666e836aa32d5a13e1e2e9220507ac?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Tinybird","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":"Tinybird","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/Tinybird-Avatar","url":"https://daily.dev/sources/tinybird"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"architecture,observability,clickhouse,tinybird","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Tinybird","item":"https://daily.dev/sources/tinybird"},{"@type":"ListItem","position":3,"name":"Cluster Management: keep the controls, skip the cluster ops"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/cluster-management-keep-the-controls-skip-the-cluster-ops-5ctirxu5a#faq","mainEntity":[{"@type":"Question","name":"How do read and write weights distribute traffic across ClickHouse replicas in Tinybird's Cluster Management?","acceptedAnswer":{"@type":"Answer","text":"Weights use proportional, weighted round-robin distribution, so a read weight of 2 receives twice the query traffic of a read weight of 1. Reader weights control query traffic, writer weights control Events API and Kafka ingestion, and copyjob weights control Copy Pipe jobs; each weight ranges from 0 to 65535, and at least one replica must have a non-zero read weight and one a non-zero write weight. Teams tuning ClickHouse replica traffic can find this kind of operational detail curated on daily.dev."}},{"@type":"Question","name":"How can I resize a ClickHouse cluster on Tinybird without downtime?","acceptedAnswer":{"@type":"Answer","text":"Use an add, reroute, verify, remove workflow: add replicas at the target CPU and memory size, wait for them to become available, rebalance read and write traffic onto them, verify query latency and ingestion throughput, then remove the old replicas. Old replicas keep serving until removed, so the cluster stays available, though both old and new capacity are billed during the overlap, which typically lasts a few minutes. Engineers planning zero-downtime database resizing can track workflows like this on daily.dev."}},{"@type":"Question","name":"How do I query CPU usage metrics for a Tinybird Dedicated ClickHouse cluster?","acceptedAnswer":{"@type":"Answer","text":"Query the organization.metrics_logs Service Data Source, which receives new records every minute and is free to query. An example selects cluster, host, and average CPU usage percentage over the last 5 minutes grouped by cluster and host; the same source also exposes memory, executing queries, system load, and replication queue metrics, with organization.endpoint_errors covering API errors. Developers monitoring database cluster health can follow practical observability approaches like this on daily.dev."}}]} ```