OpenSearchCon Europe 2026
as the organisers put it
OpenSearchCon is the annual conference uniting the OpenSearch community. Users, admins, and developers connect, solve real-world problems, and explore the future of search.
| title | author | length |
|---|---|---|
| Keynote: Opening Remarks | Bianca Lewis | 13m |
OpenSearch is five years old and adoption data shows more than half of companies have adopted or are POCing vector databases, with vendors rapidly adding… | ||
| Keynote: OpenSearch in the Agentic Era | Carl Meadows | 25m |
OpenSearch claims vectors will grow ~41.9% CAGR to 2030 and ships concrete agentic tooling: Launchpad agent skill that auto‑inspects a local movies JSON… | ||
| Keynote: The Next Generation of Enterprise Search: Intent‑Aware, Hybrid, and Open | Dom Couldwell | 28m |
Argues enterprise search needs intent-aware systems built around query intent, journey intent, product intent, mixing lexical and semantic methods rather than… | ||
| Beyond Hot Spots: Revolutionising OpenSearch Performance Through Heat-Aware Operations | G. Bafna, A. Bandejiya | 37m |
Proposes a heat-aware allocator that scores each shard as a five-dimensional vector (CPU, JVM, disk I/O, disk usage, OS requests), sampled every 5s and… | ||
| Running OpenSearch in the Real World: Migration, Mistakes and Maturity | Tom Snuverink, Marco Arts | 25m |
Runs OpenSearch on-prem on VMs as a Splunk replacement and then as an MDR logging platform ingesting 40GB to 1.2TB/day per customer, deployed with Terraform… | ||
| A Brief Guide to Secure Agentic Knowledge Retrieval | Christian Gross | 21m |
Explains how OpenSearch indexes enforce source permissions for agentic RAG by using admin connectors instead of per‑user tokens, storing an ACL field with… | ||
| Alert Fatigue to Action: Observability as the Backbone of Practical AIOps with OpenSearch | S. Pampaloni | 27m |
Argues that unifying logs, metrics and traces into OpenSearch plus OpenTelemetry and Prometheus turns observability into an AIOps substrate that actually… | ||
| Beyond Benchmarks: Real-World AI Search Performance in OpenSearch (Lucene Vs Faiss Vs JVector) | F. Rejon Barrera | 33m |
Compares Lucene HNSW, Faiss and jVector inside OpenSearch using a real client index cloned into blue/green experiments, 100 queries on ~100k vectors (K=100)… | ||
| From Embeddings To Index: A Practitioner's Guide To Domain-Adapted Neural Search | Samuel Herman | 41m |
Off-the-shelf embeddings lose most relevance on private jargon, costing 40–60% search relevance versus 1–5% ANN indexing loss; the practical pipeline is… | ||
| Building an Inclusive, Diverse Community From the Ground Up | Kassian Wren | 13m |
Lays out concrete steps: run documentation through accessibility tests, add good first issue tags and translations, create ambassador programs and safety… | ||
| Enabling Secure, Fine-Grained Resource Sharing for Team Collaboration in OpenSearch | D. Chanpura, C. Perkins | 38m |
OpenSearch adds an owner-driven sharing layer to the Security plugin called Resource Sharing Framework that replaces ad-hoc “filter by backend role” and tenant… | ||
| Vendor-Agnostic Analytics: Querying Open Columnar Formats With OpenSearch | Bharathwaj G, A. Bandejiya | 19m |
OpenSearch adds a pluggable composite engine that natively queries open columnar files (tested with Parquet) using Apache Arrow in-memory layout and DataFusion… | ||
| Integrating Docling With OpenSearch for Advanced RAG and Agentic Applications | Cesar Berrospi Ramis | 48m |
Docling converts PDFs, DOCX, PPTX and images into a unified Docling document that preserves layout, tables, figures and metadata for RAG; it ships a hybrid… | ||
| Accelerating Aggregations with Lucene-Aware Execution in OpenSearch | Ankit Jain | 39m |
Claim: replacing per-document JVM-heavy collectors with Lucene bulk primitives plus OpenSearch batching reduces aggregation CPU overhead dramatically… | ||
| Getting Started With OpenSearch Development | Andrew Ross, Saurabh Singh | 45m |
OpenSearch repository totals about 1.5 million lines of Java, builds with Gradle and requires JDK 21; a local dev server starts with gradlew run and exposes… | ||
| Building RAG With OpenSearch ML Plugins: From PDFs To Voice-Enabled Search | Kushagra Sharma | 21m |
Shows how to deploy a sentence-transformer (multilingual MiniLM converted to torchscript) onto OpenSearch via the ML Commons plugin, extract text from PDFs… | ||
| Ingesting Observability Data: OpenTelemetry Collector or Data Prepper? | Karsten Schnitter | 24m |
Compares OpenTelemetry Collector and Data Prepper with concrete tradeoffs: Collector is Go-based, low-memory and ideal as a sidecar/daemonset with low memory… | ||
| Late Interaction Retrieval in Practice: Implications for OpenSearch-Based Search Systems | F. L. de la Rosa | 18m |
Late interaction retrieval is practical as a rescoring stage in OpenSearch: documents store one pooled canonical vector for ANN indexing and separate token… | ||
| Apache Lucene Vector Search Update | Michael Sokolov | 42m |
Survey of recent Lucene vector work that lists concrete wins and how to get them: scalar quantization now supports 1-bit documents / 4-bit queries and an… | ||
| Distributed OpenSearch Monitoring at Scale With Apache NiFi and MiNiFi Agents | Vincenzo Lombardo | 18m |
Architecture using Apache NiFi and MiNiFi deploys a MiNiFi agent per node with a custom NodeStatsProcessor to collect CPU/heap/JVM/I/O/thread‑pool metrics… | ||
| Agentic Relevance Tuning With OpenSearch: Faster, Easier, and Higher Quality Search at Scale | B. Mohammed, D. Wrigley | 40m |
A new end-to-end system integrates OpenSearch User Behavior Insights, Search Relevance Workbench and an OpenSearch Agent Server to run a multi-agent pipeline… | ||
| Operating OpenSearch at Scale With Kubernetes Operators: Lessons Learned and Community Insights | C. Dinse, P. Godithi | 39m |
Runs OpenSearch at scale for 17,000 instances using a homegrown Kubernetes operator that models flexible node groups and uses StatefulSets with… | ||
| Vectors Vs. Hallucinations: OpenSearch's GenAI Survival Kit | Laysa Uchoa | 17m |
Shows concrete RAG pipelines that reduce hallucinations by 80%+ error reduction using OpenSearch vector storage, the k‑NN plugin and hybrid BM25+vector fusion… | ||
| Unconference Session: Dedicated Coordination Nodes - Why? When? How? | Aswath Srinivasan | 15m |
Amazon OpenSearch Service adds dedicated coordination nodes (DCN) to isolate coordination work like shard routing, aggregations, hybrid/vector normalization… | ||
| Unconference Session: Pains of Multi-Tenancy | David Riepl | 12m |
RISE runs an internal OpenSearch platform serving many projects and faces concrete pains: tenants duplicate log volume overnight forcing emergency capacity… | ||
| Unconference Session: Using Open Source as Your First Job | Lola Egherman | 15m |
A free mentorship program called CodeDay Labs helps about 1,000 students a year make their first open source contribution and then connects them to industry… | ||
| Unconference Session: We Should Have a Hackathon | Eric Pugh, Charlie Hull | 08m |
Proposes an official OpenSearch hackathon to turn users into contributors by getting OpenSearch running on laptops, with practical on‑the‑ground help, records… | ||
| Unconference Session: We Need Query Rewriting | Eric Pugh | 15m |
Argues that OpenSearch lacks a business-facing query rewriting layer and shows Quirky, an open source rules engine used at massive scale (Walmart Labs) to… | ||
| Unconference Session: A Neat Trick for Fuzzy Search | Charlie Hull | 13m |
Indexed 100 sports-product records from a demo dataset using Python and a free Ivan free tier OpenSearch cluster, then enables fuzzy matching with… | ||
| Unconference Session: From RFP to Fips | Kai Sternad | 10m |
OpenSearch claims FIPS compliance since 3.4 after an iterative community-driven effort sponsored by SAP, replacing noncompliant crypto with a Bouncy Castle… | ||
| Unconference Session: OpenSearch 4.0: What? When? Why? | Andrew Ross | 15m |
Proposes treating OpenSearch 4.0 as a policy decision not just a feature bump, separating three real breaking-change classes: API compatibility (avoid removing… | ||
| Unconference Session: What Alternatives to KeepHQ? | Stefano Pampaloni | 14m |
KeepHQ is an alert manager that connects to 100–120 providers to dedupe, group and create correlated incidents; Elastic’s acquisition slows upstream… | ||
| OpenSearchCon Europe 2026 Highlight Video | 01m | |
Highlight reel shows conference atmosphere, packed sessions and hallway conversations and demos of agentic features that claim to build OpenSearch projects… | ||
| OpenSearchCon Europe 2026 Highlight Video | 01m | |
A three‑minute highlight reel compiles short clips from sessions and hallway conversations and focuses on concrete advances: a set of agentic features that… | ||
| Keynote: Bianca Lewis, Executive Director, OpenSearch Software Foundation, The Linux Foundation | Bianca Lewis | 18m |
OpenSearch Foundation reports concrete growth metrics — 1.5 billion downloads, ~100 ecosystem vendors, ~150 GitHub repos, 4,000 Slack members and 500+… | ||
| Sponsored Keynote: AI-Driven Deployment and Monitoring Across Federated Cloud-Edge Sites | G. Conti, B. Ginestra | 17m |
Demonstrates turning natural language prompts into executable deployment requests that run across a federated cloud‑edge continuum spanning three Italian… | ||
| The Forbidden Index: How We Turned OpenSearch Into a Privacy-Preserving Superhero | U. Mishra, A. Khanna | 36m |
This project converts OpenSearch into a privacy-aware analytics engine by adding an PII auto-redaction plugin that scans ingest pipelines (regex + NLP)… | ||
| The Lucene Scalpel: Tuning Low-Level Segments for Multi-Lingual Concurrency | Divyanshu Mishra | 21m |
Shows concrete Lucene settings and measured gains: uses OpenSearch with per-field analyzers (SmartCN for Chinese, Arabic analyzer, English analyzers), adjusts… | ||
| Writable Warm: Breaking the Read-Only Barrier in OpenSearch Tier... | S. Kumawat, H. Kaushik, S. Sahu | 24m |
Writable Warm makes warm-tier indices writable by routing Lucene I/O through a composite directory that unifies local disk and remote blob stores… | ||
| Fluent Bit & OpenSearch: The Power Couple of Open Source Observability | Patrick Stephens | 31m |
Fluent Bit is presented as a low‑resource, high‑performance edge collector that forwards enriched telemetry into OpenSearch using a native OpenSearch output… | ||
| Building a Modern Search Engine: The Road Ahead | Eric Pugh | 44m |
Roadmap for 2026 commits OpenSearch to attract data scientists and AI engineers by compressing complexity so hybrid search and rerankers work out of the box… | ||
| Teaching OpenSearch To Read YAML: Finding Broken Infrastructure Before It Breaks | Unnati Mishra | 18m |
A pipeline converts committed Kubernetes YAML into structured JSON via a custom parser that mimics the API server, resolves label/selector joins into arrays… | ||
| Upcoming Changes for the OpenSearch Index Authorization Mechanisms | Nils Bandener | 40m |
OpenSearch switches from the old all‑or‑nothing index authorization to a safer, request‑level resolution that strips unauthorized indices from index… | ||
| Road To Maintainership: Shaping OpenSearch's Future | Varun Bansal | 19m |
Practical playbook showing which repo areas to join, how triage meetings are used to prioritize work, and what tests and artifacts reviewers actually expect —… | ||
| Fine-tuning Neural Sparse Model for Domain Specific Data From an Existing OpenSearch Index | A. Srinivasan, C. Pelvet | 41m |
Fine-tunes sparse encoders using an OpenSearch index, Hugging Face/BERT variants and open-source tooling, exporting models as TorchScript and deploying them… | ||
| Scaling Welfare With Open Source: How NAV Migrated To Managed OpenSearch | D. Kan, H.K. Flaatten | 41m |
Norway’s NAV migrates a 60TB logging footprint from on‑prem Elastic to managed OpenSearch with a zero‑downtime dual‑write migration, moving hundreds of GB/day… | ||
| From Logs To Signals: Rethinking System Monitoring With OpenSearch | Shruti Chaturvedi | 18m |
Argues that structured JSON logs become first-class signals indexed into OpenSearch so sliding‑window aggregation queries continuously evaluate system state… | ||
| Caching for Speed: Practical Performance Optimization in OpenSearch & Lucene | S. Upadhyaya, A. Jain | 30m |
Shows concrete changes: moving FST off-heap and similar structures to native memory to avoid JVM GC pressure, and refactoring the Lucene query cache into ~16… | ||
| Unified Performance Observability in OpenSearch: Integrating Workload Management and Query Insights | L. Chen, E. Guo | 20m |
Query Insights records latency CPU and memory by search phase and treats WLM group info as a first class attribute so queries are filterable and grouped by… | ||
| Sharding 2.0: Unlocking Infinite Search Scalability in OpenSearch | Vikas Bansal, Paras Jain | 31m |
OpenSearch adds an in-place shard split that turns a static shard count into a dynamic root→child hierarchy with mutually exclusive hash ranges and a… | ||
| Boosting OpenSearch Performance: Lucene Bulk Collection and gRPC Search in Action | A.M. Kolarkunnu, C. Rolo | 30m |
Lucene’s new bulk collection APIs (LongValues, stream.intoArray, collectRange) batch doc handling so JVM can inline and avoid virtual dispatch, producing… | ||
| The OSS-ential Understanding of Community Health and Metrics | Ijeoma Onwuka | 27m |
Open source health is not star counts but measurable risk and resilience: Linux Foundation Tide cites 60% unpaid maintainers and ~60% who consider quitting… | ||
| Unlocking Next-gen Semantic Search With OpenSearch 3.x | Pietro Mele, Lucian Precup | 48m |
OpenSearch 3.x demonstrates concrete recipes for billion‑scale vectors: GPU indexing via Cagra, Faiss and JVector engines, and HNSW/IVF ANN with Radial Search… | ||
| Using Native CPU Vectorized Instructions for 60% Speedup on Lucene's KNN Vector Search | S. Chaudhary | 40m |
Lucene’s dot‑product hotspot gets rewritten in native SIMD C and delivers a production 60% latency reduction on ARM, with microbenchmarks showing up to 10–13x… | ||
| Keynote: Closing Remarks | Kris Freedain, Patti Juric | 13m |
Foundation announces creation in September 2024 and grows from 12 founding members to multiple new general and premier partners; GitHub contributions and… | ||
| nothing matches | ||