https://qdrant.tech/documentation/production-operations/Recent content in Production & Operations on Qdrant - Vector Search EngineHugoen-usinfo@qdrant.tech (Andrey Vasnetsov)info@qdrant.tech (Andrey Vasnetsov)https://qdrant.tech/documentation/production-operations/memory-tiers/Fri, 28 Aug 2026 10:00:00 +0200info@qdrant.tech (Andrey Vasnetsov)https://qdrant.tech/documentation/production-operations/memory-tiers/<h1 id="memory-tiers-in-qdrant-what-to-use-and-when">Memory Tiers in Qdrant: What to Use and When</h1> <p>A growing vector collection eventually outgrows the RAM it started with: Qdrant handles that by letting you assign dense vectors, the HNSW graph, quantized vectors, payloads, and payload indexes each to whichever memory tier that structure supports, instead of forcing one RAM-versus-disk trade-off onto the whole collection.</p> <p>This article will give you practical guidance over which combination of tiers and quantization to reach for at each stage of a collection&rsquo;s growth, and the reasons behind the choice.</p>https://qdrant.tech/documentation/production-operations/bulk-data-import/Tue, 14 Jul 2026 00:00:00 +0000info@qdrant.tech (Andrey Vasnetsov)https://qdrant.tech/documentation/production-operations/bulk-data-import/<h1 id="bulk-uploading-data-to-qdrant">Bulk Uploading Data to Qdrant</h1> <h2 id="why-bulk-uploading-matters">Why Bulk Uploading Matters</h2> <p>When you start using Qdrant at scale, one of the first challenges you may run into is uploading large amounts of data efficiently. Small uploads are usually straightforward, but bulk ingestion introduces a different set of concerns. As millions of vectors, payloads, and indexes are written into a collection, the system has to manage memory usage, disk writes, background optimization, and search availability at the same time.</p>https://qdrant.tech/documentation/production-operations/multitenant-search/Tue, 06 Feb 2024 13:21:00 +0000info@qdrant.tech (Andrey Vasnetsov)https://qdrant.tech/documentation/production-operations/multitenant-search/<h1 id="scaling-your-machine-learning-setup-the-power-of-multitenancy-and-custom-sharding-in-qdrant">Scaling Your Machine Learning Setup: The Power of Multitenancy and Custom Sharding in Qdrant</h1> <p>We are seeing the topics of <a href="https://qdrant.tech/documentation/manage-data/multitenancy/">multitenancy</a> and <a href="https://qdrant.tech/documentation/scaling/distributed_deployment/#sharding">distributed deployment</a> pop-up daily on our <a href="https://qdrant.to/discord" target="_blank" rel="noopener nofollow">Discord support channel</a>. This tells us that many of you are looking to scale Qdrant along with the rest of your machine learning setup.</p> <p>Whether you are building a bank fraud-detection system, <a href="https://qdrant.tech/articles/what-is-rag-in-ai/" target="_blank" rel="noopener nofollow">RAG</a> for e-commerce, or services for the federal government - you will need to leverage a multitenant architecture to scale your product. In the world of SaaS and enterprise apps, this setup is the norm. It will considerably increase your application&rsquo;s performance and lower your hosting costs.</p>