From e565b19687c5dcb2a5624ae8cfb01d7161dd546f Mon Sep 17 00:00:00 2001 From: Maddie Duhon Date: Fri, 7 Mar 2025 09:08:40 -0500 Subject: [PATCH] Removed-Milvus Per Arun's request, removed the 2 mentions of Milvus --- qdrant-landing/content/blog/case-study-deutsche-telekom.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/qdrant-landing/content/blog/case-study-deutsche-telekom.md b/qdrant-landing/content/blog/case-study-deutsche-telekom.md index 519649452..9a4c2a02c 100644 --- a/qdrant-landing/content/blog/case-study-deutsche-telekom.md +++ b/qdrant-landing/content/blog/case-study-deutsche-telekom.md @@ -69,7 +69,7 @@ LMOS architecture powering AI agent collaboration and lifecycle management in a ### Why Qdrant? Finding the Right Vector Database for LMOS -When Deutsche Telekom began searching for a scalable, high-performance vector database, they faced operational challenges with their initial choice, Milvus. Seeking a solution better suited to their PaaS-first approach and multitenancy requirements, they evaluated alternatives, and [Qdrant](https://qdrant.tech/qdrant-vector-database/) quickly stood out. +When Deutsche Telekom began searching for a scalable, high-performance vector database, they faced operational challenges with their initial choice. Seeking a solution better suited to their PaaS-first approach and multitenancy requirements, they evaluated alternatives, and [Qdrant](https://qdrant.tech/qdrant-vector-database/) quickly stood out. "I was looking for open-source components with deep technical expertise behind them," Arun recalls. "I looked at Qdrant and immediately loved the simplicity, [Rust-based efficiency](https://qdrant.tech/articles/why-rust/), and [memory management capabilities](https://qdrant.tech/articles/memory-consumption/). These guys knew what they were doing." @@ -84,7 +84,7 @@ Deutsche Telekom's engineers also cited several standout features that made Qdra 2. **Developer experience**—libraries, multi-language clients, and cross-framework support make integrations seamless. 3. **WebUI & Collection Visualization**—engineers found Qdrant's [built-in collection visualization](https://qdrant.tech/documentation/web-ui/) tools highly useful. -As part of their evaluation, Deutsche Telekom engineers compared multiple solutions, weighing operational simplicity and reliability. One engineer summarized their findings: "Qdrant has way fewer components. Milvus required Kafka, Zookeeper, and only had a hot standby for its index and query nodes. If you rescale Milvus, you get downtime. Qdrant stays up." +As part of their evaluation, Deutsche Telekom engineers compared multiple solutions, weighing operational simplicity and reliability. One engineer summarized their findings: "Qdrant has way fewer components, compared to the previous solution that required required Kafka, Zookeeper, and only had a hot standby for its index and query nodes. If you rescaled it, you get downtime. Qdrant stays up." ### Scaling AI at Deutsche Telekom & The Future of LMOS