[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-qdrant-qdrant-horizontal-scaling":3,"mdc--ut9iow-key":48,"related-repo-qdrant-qdrant-horizontal-scaling":225,"related-org-qdrant-qdrant-horizontal-scaling":318},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":43,"sourceUrl":46,"mdContent":47},"qdrant-horizontal-scaling","guide Qdrant horizontal scaling decisions","Diagnoses and guides Qdrant horizontal scaling decisions. Use when someone asks 'vertical or horizontal?', 'how many nodes?', 'how many shards?', 'how to add nodes', 'resharding', 'data doesn't fit', or 'need more capacity'. Also use when data growth outpaces current deployment.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"qdrant","Qdrant","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fqdrant.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"Architecture","architecture","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Scaling","scaling",{"name":21,"slug":22,"type":15},"Database","database",197,"https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills","2026-07-19T05:38:32.348551",null,23,[29,30,31,32,33,34,35,36,37,38,8,39,19,40,41,42],"agent-skills","ai-agents","claude-code","codex","cursor","embeddings","hybrid-search","monitoring","multitenancy","performance","quantization","search-quality","vector-database","vector-search",{"repoUrl":24,"stars":23,"forks":27,"topics":44,"description":45},[29,30,31,32,33,34,35,36,37,38,8,39,19,40,41,42],"Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java","https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fqdrant-scaling\u002Fscaling-data-volume\u002Fhorizontal-scaling","---\nname: qdrant-horizontal-scaling\ndescription: \"Diagnoses and guides Qdrant horizontal scaling decisions. Use when someone asks 'vertical or horizontal?', 'how many nodes?', 'how many shards?', 'how to add nodes', 'resharding', 'data doesn't fit', or 'need more capacity'. Also use when data growth outpaces current deployment.\"\n---\n\n# What to Do When Qdrant Needs More Capacity\n\nVertical first: simpler operations, no network overhead, good up to ~100M vectors per node depending on dimensions and quantization. Horizontal when: data exceeds single node capacity, need fault tolerance, need to isolate tenants, or IOPS-bound (more nodes = more independent IOPS).\n\n## Most basic distributed configuration\n\n- 3 nodes, 3 shards with `replication_factor: 2` for zero-downtime scaling\n\nMinimum of 3 nodes is important for consensus and fault tolerance. With 3 nodes, you can lose 1 node without downtime. With 2 nodes, losing 1 node causes downtime for collection operations.\nReplication factor of 2 means each shard has 1 replica, so you have 2 copies of data. This allows for zero-downtime scaling and maintenance. With `replication_factor: 1`, zero-downtime is not guaranteed even for point-level operations, and cluster maintenance requires downtime.\n\n## Choosing number of shards\n\nShards are the unit of data distribution. \nMore shards allows more nodes and better distribution, but adds overhead. Fewer shards reduces overhead but limits horizontal scaling.\n\nFor cluster of 3-6 nodes the recommended shard count is 6-12. \nThis allows for 2-4 shards per node, which balances distribution and overhead. \n\n## Changing number of shards\n\nUse when: shard count isn't evenly divisible by node count, causing uneven distribution, or need to rebalance.\n\nResharding is expensive and time-consuming, it should be used as a last resort if regular data distribution is not possible.\nResharding is designed to be transparent for user operations, updates and searches should still work during resharding with some small performance impact.\n\nBut resharding operation itself is time-consuming and requires to move large amounts of data between nodes.\n\n- Available in Qdrant Cloud [Resharding](https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fscaling\u002Fdistributed_deployment\u002F?s=resharding)\n- Resharding is not available for self-hosted deployments.\n\nBetter alternatives: over-provision shards initially, or spin up new cluster with correct config and migrate data.\n\n\n## What NOT to Do\n\n- Do not jump to horizontal before exhausting vertical (adds complexity for no gain)\n- Do not set `shard_number` that isn't a multiple of node count (uneven distribution)\n- Do not use `replication_factor: 1` in production if you need fault tolerance\n- Do not add nodes without rebalancing shards (use shard move API to redistribute)\n- Do not scale down RAM without load testing (cache eviction causes days-long latency incidents)\n- Do not hit the collection limit by using one collection per tenant (use payload partitioning)\n",{"data":49,"body":50},{"name":4,"description":6},{"type":51,"children":52},"root",[53,62,68,75,94,107,113,118,123,129,134,139,144,166,171,177],{"type":54,"tag":55,"props":56,"children":58},"element","h1",{"id":57},"what-to-do-when-qdrant-needs-more-capacity",[59],{"type":60,"value":61},"text","What to Do When Qdrant Needs More Capacity",{"type":54,"tag":63,"props":64,"children":65},"p",{},[66],{"type":60,"value":67},"Vertical first: simpler operations, no network overhead, good up to ~100M vectors per node depending on dimensions and quantization. Horizontal when: data exceeds single node capacity, need fault tolerance, need to isolate tenants, or IOPS-bound (more nodes = more independent IOPS).",{"type":54,"tag":69,"props":70,"children":72},"h2",{"id":71},"most-basic-distributed-configuration",[73],{"type":60,"value":74},"Most basic distributed configuration",{"type":54,"tag":76,"props":77,"children":78},"ul",{},[79],{"type":54,"tag":80,"props":81,"children":82},"li",{},[83,85,92],{"type":60,"value":84},"3 nodes, 3 shards with ",{"type":54,"tag":86,"props":87,"children":89},"code",{"className":88},[],[90],{"type":60,"value":91},"replication_factor: 2",{"type":60,"value":93}," for zero-downtime scaling",{"type":54,"tag":63,"props":95,"children":96},{},[97,99,105],{"type":60,"value":98},"Minimum of 3 nodes is important for consensus and fault tolerance. With 3 nodes, you can lose 1 node without downtime. With 2 nodes, losing 1 node causes downtime for collection operations.\nReplication factor of 2 means each shard has 1 replica, so you have 2 copies of data. This allows for zero-downtime scaling and maintenance. With ",{"type":54,"tag":86,"props":100,"children":102},{"className":101},[],[103],{"type":60,"value":104},"replication_factor: 1",{"type":60,"value":106},", zero-downtime is not guaranteed even for point-level operations, and cluster maintenance requires downtime.",{"type":54,"tag":69,"props":108,"children":110},{"id":109},"choosing-number-of-shards",[111],{"type":60,"value":112},"Choosing number of shards",{"type":54,"tag":63,"props":114,"children":115},{},[116],{"type":60,"value":117},"Shards are the unit of data distribution.\nMore shards allows more nodes and better distribution, but adds overhead. Fewer shards reduces overhead but limits horizontal scaling.",{"type":54,"tag":63,"props":119,"children":120},{},[121],{"type":60,"value":122},"For cluster of 3-6 nodes the recommended shard count is 6-12.\nThis allows for 2-4 shards per node, which balances distribution and overhead.",{"type":54,"tag":69,"props":124,"children":126},{"id":125},"changing-number-of-shards",[127],{"type":60,"value":128},"Changing number of shards",{"type":54,"tag":63,"props":130,"children":131},{},[132],{"type":60,"value":133},"Use when: shard count isn't evenly divisible by node count, causing uneven distribution, or need to rebalance.",{"type":54,"tag":63,"props":135,"children":136},{},[137],{"type":60,"value":138},"Resharding is expensive and time-consuming, it should be used as a last resort if regular data distribution is not possible.\nResharding is designed to be transparent for user operations, updates and searches should still work during resharding with some small performance impact.",{"type":54,"tag":63,"props":140,"children":141},{},[142],{"type":60,"value":143},"But resharding operation itself is time-consuming and requires to move large amounts of data between nodes.",{"type":54,"tag":76,"props":145,"children":146},{},[147,161],{"type":54,"tag":80,"props":148,"children":149},{},[150,152],{"type":60,"value":151},"Available in Qdrant Cloud ",{"type":54,"tag":153,"props":154,"children":158},"a",{"href":155,"rel":156},"https:\u002F\u002Fskills.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fscaling\u002Fdistributed_deployment\u002F?s=resharding",[157],"nofollow",[159],{"type":60,"value":160},"Resharding",{"type":54,"tag":80,"props":162,"children":163},{},[164],{"type":60,"value":165},"Resharding is not available for self-hosted deployments.",{"type":54,"tag":63,"props":167,"children":168},{},[169],{"type":60,"value":170},"Better alternatives: over-provision shards initially, or spin up new cluster with correct config and migrate data.",{"type":54,"tag":69,"props":172,"children":174},{"id":173},"what-not-to-do",[175],{"type":60,"value":176},"What NOT to Do",{"type":54,"tag":76,"props":178,"children":179},{},[180,185,198,210,215,220],{"type":54,"tag":80,"props":181,"children":182},{},[183],{"type":60,"value":184},"Do not jump to horizontal before exhausting vertical (adds complexity for no gain)",{"type":54,"tag":80,"props":186,"children":187},{},[188,190,196],{"type":60,"value":189},"Do not set ",{"type":54,"tag":86,"props":191,"children":193},{"className":192},[],[194],{"type":60,"value":195},"shard_number",{"type":60,"value":197}," that isn't a multiple of node count (uneven distribution)",{"type":54,"tag":80,"props":199,"children":200},{},[201,203,208],{"type":60,"value":202},"Do not use ",{"type":54,"tag":86,"props":204,"children":206},{"className":205},[],[207],{"type":60,"value":104},{"type":60,"value":209}," in production if you need fault tolerance",{"type":54,"tag":80,"props":211,"children":212},{},[213],{"type":60,"value":214},"Do not add nodes without rebalancing shards (use shard move API to redistribute)",{"type":54,"tag":80,"props":216,"children":217},{},[218],{"type":60,"value":219},"Do not scale down RAM without load testing (cache eviction causes days-long latency incidents)",{"type":54,"tag":80,"props":221,"children":222},{},[223],{"type":60,"value":224},"Do not hit the collection limit by using one collection per tenant (use payload partitioning)",{"items":226,"total":317},[227,242,257,272,287,294,307],{"slug":228,"name":228,"fn":229,"description":230,"org":231,"tags":232,"stars":23,"repoUrl":24,"updatedAt":241},"qdrant-advisor","diagnose and troubleshoot Qdrant deployments","Diagnose, troubleshoot, and advise on any Qdrant deployment by loading the latest official Qdrant skills live from skills.qdrant.tech. Use this whenever someone raises a Qdrant problem or question — slow or degraded search, high or growing memory \u002F OOM crashes, optimizer stuck or slow, indexing slowness, scaling and sharding decisions (node count, QPS, latency, multitenancy, vertical vs horizontal), poor or irrelevant search results, hybrid search and reranking, embedding-model migration, version upgrades and compatibility, monitoring and observability (Prometheus, Grafana, health checks, \u002Fmetrics, \u002Ftelemetry), deployment choices (local, Docker, self-hosted, Qdrant Cloud, embedded), or client-SDK questions (Python, TypeScript, Rust, Go, .NET, Java). Trigger especially when the context is clearly a Qdrant cluster, collection, or vector-search deployment. Always prefer this skill over answering from memory: it pulls current, authoritative guidance and only the relevant context.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[233,234,237,240],{"name":21,"slug":22,"type":15},{"name":235,"slug":236,"type":15},"Debugging","debugging",{"name":238,"slug":239,"type":15},"Operations","operations",{"name":9,"slug":8,"type":15},"2026-07-16T06:02:56.675293",{"slug":243,"name":243,"fn":244,"description":245,"org":246,"tags":247,"stars":23,"repoUrl":24,"updatedAt":256},"qdrant-clients-sdk","integrate Qdrant client SDKs","Qdrant provides client SDKs for various programming languages, allowing easy integration with Qdrant deployments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[248,251,252,253],{"name":249,"slug":250,"type":15},"API Development","api-development",{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":254,"slug":255,"type":15},"SDK","sdk","2026-07-16T05:59:57.799789",{"slug":258,"name":258,"fn":259,"description":260,"org":261,"tags":262,"stars":23,"repoUrl":24,"updatedAt":271},"qdrant-deployment-options","select Qdrant deployment options","Guides Qdrant deployment selection. Use when someone asks 'how to deploy Qdrant', 'Docker vs Cloud', 'local mode', 'embedded Qdrant', 'Qdrant EDGE', 'which deployment option', 'self-hosted vs cloud', or 'need lowest latency deployment'. Also use when choosing between deployment types for a new project.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[263,264,267,270],{"name":21,"slug":22,"type":15},{"name":265,"slug":266,"type":15},"Deployment","deployment",{"name":268,"slug":269,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-19T05:38:33.324265",{"slug":273,"name":273,"fn":274,"description":275,"org":276,"tags":277,"stars":23,"repoUrl":24,"updatedAt":286},"qdrant-edge","build applications with Qdrant Edge","Guides building on Qdrant Edge, the embedded in-process shard. Use when someone asks 'how to sync Edge with the server', 'keep a local shard in sync with Qdrant Cloud', 'BM25 or keyword search on Edge', 'hybrid search on Edge', 'embeddings on device', 'Edge snapshots', 'apply a partial snapshot', 'why is my Edge search empty after inserts', or is writing custom sync, BM25, or fusion code against qdrant-edge. Also use when deciding what Edge ships built-in versus what you must implement.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[278,279,282,283],{"name":21,"slug":22,"type":15},{"name":280,"slug":281,"type":15},"Edge","edge",{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},"Search","search","2026-07-16T06:01:19.179105",{"slug":4,"name":4,"fn":5,"description":6,"org":288,"tags":289,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[290,291,292,293],{"name":13,"slug":14,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},{"slug":295,"name":295,"fn":296,"description":297,"org":298,"tags":299,"stars":23,"repoUrl":24,"updatedAt":306},"qdrant-hybrid-search","implement hybrid search in Qdrant","Explains hybrid search in Qdrant. Use when someone asks 'how do I setup hybrid search?', 'how to combine keyword and semantic search?', 'sparse plus dense vectors?', 'missing keyword matches', 'how to combine results from multiple searches?' and 'combining multiple representations'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[300,303,304,305],{"name":301,"slug":302,"type":15},"AI Infrastructure","ai-infrastructure",{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},"2026-07-16T06:00:12.328122",{"slug":308,"name":308,"fn":309,"description":310,"org":311,"tags":312,"stars":23,"repoUrl":24,"updatedAt":316},"qdrant-hybrid-search-combining","combine hybrid search scores in Qdrant","Fusing scores from multiple searches into a single ranked result (RRF, DBSF, custom fusion). Use when someone asks 'RRF or DBSF?', 'how to combine sparse and dense', 'how to combine scores from multiple searches?', 'custom fusion', or 'fusion is not producing good results'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[313,314,315],{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},"2026-07-16T06:01:49.401142",30,{"items":319,"total":317},[320,327,334,341,348,355,362,368,379,391,401,412],{"slug":228,"name":228,"fn":229,"description":230,"org":321,"tags":322,"stars":23,"repoUrl":24,"updatedAt":241},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[323,324,325,326],{"name":21,"slug":22,"type":15},{"name":235,"slug":236,"type":15},{"name":238,"slug":239,"type":15},{"name":9,"slug":8,"type":15},{"slug":243,"name":243,"fn":244,"description":245,"org":328,"tags":329,"stars":23,"repoUrl":24,"updatedAt":256},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[330,331,332,333],{"name":249,"slug":250,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":254,"slug":255,"type":15},{"slug":258,"name":258,"fn":259,"description":260,"org":335,"tags":336,"stars":23,"repoUrl":24,"updatedAt":271},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[337,338,339,340],{"name":21,"slug":22,"type":15},{"name":265,"slug":266,"type":15},{"name":268,"slug":269,"type":15},{"name":9,"slug":8,"type":15},{"slug":273,"name":273,"fn":274,"description":275,"org":342,"tags":343,"stars":23,"repoUrl":24,"updatedAt":286},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[344,345,346,347],{"name":21,"slug":22,"type":15},{"name":280,"slug":281,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},{"slug":4,"name":4,"fn":5,"description":6,"org":349,"tags":350,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[351,352,353,354],{"name":13,"slug":14,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},{"slug":295,"name":295,"fn":296,"description":297,"org":356,"tags":357,"stars":23,"repoUrl":24,"updatedAt":306},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[358,359,360,361],{"name":301,"slug":302,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},{"slug":308,"name":308,"fn":309,"description":310,"org":363,"tags":364,"stars":23,"repoUrl":24,"updatedAt":316},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[365,366,367],{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},{"slug":369,"name":369,"fn":370,"description":371,"org":372,"tags":373,"stars":23,"repoUrl":24,"updatedAt":378},"qdrant-hybrid-search-prefetches","configure hybrid search with prefetch queries","Constructing prefetch queries for hybrid retrieval, including sparse\u002Fdense and multi-field setups, and choosing a sparse embedding model. Use when someone asks 'dense and sparse in one search?', 'how to combine multiple fields for retrieval?', 'payloads or sparse vectors for lexical?', 'which sparse embedding model to use?', or 'BM25 vs SPLADE?'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[374,375,376,377],{"name":249,"slug":250,"type":15},{"name":21,"slug":22,"type":15},{"name":9,"slug":8,"type":15},{"name":284,"slug":285,"type":15},"2026-07-16T06:03:02.305907",{"slug":380,"name":380,"fn":381,"description":382,"org":383,"tags":384,"stars":23,"repoUrl":24,"updatedAt":390},"qdrant-indexing-performance-optimization","optimize Qdrant indexing and data ingestion","Diagnoses and fixes slow Qdrant indexing and data ingestion. Use when someone reports 'uploads are slow', 'indexing takes forever', 'optimizer is stuck', 'HNSW build time too long', or 'data uploaded but search is bad'. Also use when optimizer status shows errors, segments won't merge, or indexing threshold questions arise.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[385,386,387,389],{"name":21,"slug":22,"type":15},{"name":235,"slug":236,"type":15},{"name":388,"slug":38,"type":15},"Performance",{"name":9,"slug":8,"type":15},"2026-07-19T05:38:36.321999",{"slug":392,"name":392,"fn":393,"description":394,"org":395,"tags":396,"stars":23,"repoUrl":24,"updatedAt":400},"qdrant-memory-usage-optimization","optimize Qdrant memory usage","Diagnoses and reduces Qdrant memory usage. Use when someone reports 'memory too high', 'RAM keeps growing', 'node crashed', 'out of memory', 'memory leak', or asks 'why is memory usage so high?', 'how to reduce RAM?'. Also use when memory doesn't match calculations, quantization didn't help, or nodes crash during recovery.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[397,398,399],{"name":235,"slug":236,"type":15},{"name":388,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:15.357243",{"slug":402,"name":402,"fn":403,"description":404,"org":405,"tags":406,"stars":23,"repoUrl":24,"updatedAt":411},"qdrant-minimize-latency","optimize Qdrant query latency","Guides Qdrant query latency optimization. Use when someone asks 'search is slow', 'how to reduce latency', 'p99 is too high', 'tail latency', 'single query too slow', 'how to make search faster', or 'latency spikes'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[407,408,409,410],{"name":21,"slug":22,"type":15},{"name":235,"slug":236,"type":15},{"name":388,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T05:59:17.325839",{"slug":413,"name":413,"fn":414,"description":415,"org":416,"tags":417,"stars":23,"repoUrl":24,"updatedAt":424},"qdrant-model-migration","migrate embedding models in Qdrant","Guides embedding model migration in Qdrant without downtime. Use when someone asks 'how to switch embedding models', 'how to migrate vectors', 'how to update to a new model', 'zero-downtime model change', 'how to re-embed my data', or 'can I use two models at once'. Also use when upgrading model dimensions, switching providers, or A\u002FB testing models.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[418,419,422,423],{"name":21,"slug":22,"type":15},{"name":420,"slug":421,"type":15},"Migration","migration",{"name":388,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:48.841055"]