OpenSearchCon Europe 2025

In this session, Uber, a premier member of the OpenSearch Foundation, will share how the unique demands of their low-latency, high-throughput search platform have driven impactful contributions to the OpenSearch project.

The discussion will explore Uber’s collaboration with the OpenSearch community to integrate critical learnings and innovations from their own search platform. Key highlights include the development of a modular, cloud-native architecture with clear separation between readers and writers, a high-performance messaging protocol for inter-service communication, and a pull-based indexing model.

These advancements, along with contributions from the broader OpenSearch community, are shaping OpenSearch into a cutting-edge, serverless open-source search engine tailored for enterprise needs. Join this session to discover how pioneering work from Uber and OpenSearch is paving the way for the next generation of high performance and scalable search technology in open source.

Leveraging OpenSearch, we propose a Cross-Reference Code Mapping System to dynamically map medical products across terminologies and regulatory frameworks. For example, it maps Paracetamol (UK) to Acetaminophen (US), including dosage details and regulatory codes. Using nested documents, the system efficiently stores and retrieves all related data. Semantic search enables queries like "pain relief tablets," returning relevant matches across terminologies.

Additionally, by incorporating Cross-Cluster Search, the system seamlessly queries datasets stored across multiple clusters, such as those dedicated to specific regions or regulatory frameworks. This ensures unified data access while maintaining localized storage and compliance with regional regulations. Cross-Cluster Search also ensures scalability, allowing new datasets or regions to be added incrementally without disrupting operations.

Powered by OpenSearch, this system improves data interoperability, enhances medication accuracy, reduces errors, and streamlines research workflows. Its scalable design adapts to the evolving needs of the global healthcare community.

The standard way for creating APIs is to first define the API contract, use that to generate classes and write your application around it. But what happens when the application is already in place and is actively being used by millions around the world and does not have an API contract?

We faced this issue when we set out to find a single source of truth for OpenSearch APIs. Finding none and realizing that we need to create one, we set out on a journey to create asuch specification.

This talk will deal with:
* Experience with Jest client and frustration with freeform request bodies as a customer
* The stages of development including ideation, PoC, RFC, choosing the technologies, gaining consensus and implementation
* setting up the project and integration framework (while ensuring frugality and complete coverage)
* Opening up the project to community contribution after initial implementation
* Adoption by the clients team
* Switch from Smithy to OpenAPI based on community discussion
and the FUTURE!

This session will guide you through the process of building a complete observability stack from the ground up using raw telemetry data ingested directly into a datastore, combining a variety of open source signals including OpenTelemetry and Prometheus formats.

We’ll start by leveraging our existing telemetry for maximum effectiveness because telemetry without action is just storage. We'll delve into the aspects of creating OLAP cubes. We’ll use this data for creating insightful visualizations with tools like Perses.

Drawing from experience with proprietary and open source observability tools, we’ll then evolve our monitoring by adding application telemetry. We’ll use automations like eBPF to fully observe a massively distributed cloud database offering.

Whether you're starting from scratch or looking to enhance your existing observability infrastructure, join me to discover practical strategies that you can implement today.

One of the really cool aspects of OpenSearch is its ability to serve as a monitoring tool for itself. Let's have a look at how to piece together a monitoring solution for OpenSearch using components from the OpenSearch ecosystem such as Data Prepper and OpenSearch alerting. We will focus on security aspects like authentication and authorization.

Any software dealing with confidential data needs to have a security solution providing authentication and authorization. And any such security solution should have proper monitoring. OpenSearch is no exception in this regard.

In this presentation, we will look how to use components like Audit Logs, Data Prepper and OpenSearch Alerting to create a small, but effective security monitoring solution.

This will give us two benefits at once:

- A security monitoring solution
- Insights into the workings of components like Alerting, etc. by looking at a real world example

In the era of vector search and semantic similarity, returning highly relevant results is only half the battle. When search results are too similar, users must wade through redundant information to find diverse perspectives. This talk introduces Maximum Marginal Relevance (MMR) implementation in OpenSearch, a powerful technique that optimally balances result relevance with diversity.

Key takeaways will include:
- Implementing MMR reranking with OpenSearch's vector search
- Optimizing performance for large-scale deployments
- Measuring and tuning diversity metrics
- Real-world applications and success patterns

In the realm of search, Machine Learning plays a pivotal role in enhancing the user experience throughout the entire lifecycle, from ingesting documents to delivering highly relevant results for user queries. This session will showcase various ML integrations of OpenSearch tailored to optimize outcomes for user queries in a retail scenario, accompanied by a live demonstration. We will explore cutting-edge techniques such as query understanding and rewriting using Large Language Models (LLMs), document enrichment, sparse, dense, and hybrid retrievers, as well as contextual re-ranking of results. As a highlight, you will learn how LLM agents and OpenSearch’s ML Search types can be integrated to build a AI shopping assistant that offers a conversational shopping experience for the users.

Scaling vector search to a billion vectors isn’t just a technical challenge—it’s a high-stakes balancing act of latency, cost, and quality. In this talk, I’ll share practical lessons learned from building a billion-scale system in OpenSearch, covering model selection, text processing, hardware essentials, and embedding strategies. But that’s not all: I’ll expose the tips, tricks, and gotchas you’ll need to navigate OpenSearch’s quirks without losing your sanity (or your wallet). Whether you’re scaling up or just getting started, this session will equip you to tame your vectors and succeed at scale.