OpenSearchCon North America 2025
2025-09-08..2025-09-10
San Jose, United States
as the organisers put it
Join the OpenSearch community at OpenSearchCon NA in San Jose, CA, Sept 8–10, 2025. Learn, connect, and explore the future of search, observability, and security.
| title | author | length |
|---|---|---|
| Highlights from the community and the conference | 01m | |
Highlights reel shows hands-on workshops on ingest, indexing, and cluster scaling alongside deep technical sessions that dig into performance tuning and… | ||
| The OpenSearch Community, Building the Future Together | Carl Meadows, Corey J. Nolet, Hariharan Gandhi | 01h 03m |
OpenSearch moves under the Linux Foundation and reports 8,800 contributions from 3,300 contributors across 400 organizations, jumping from #70 to the top 20… | ||
| Advancing Search and Observability at Uber with OpenSearch | Shanshan Song | 15m |
Uber extends OpenSearch with production-scale semantic vector search using embeddings, a two-phase pipeline (cosine retrieval to winnow billions to thousands… | ||
| OpenSearch at Atlassian Scale | Fernando Garcia Valenzuela | 13m |
Atlassian runs OpenSearch across 300+ clusters, 2,000+ data nodes, and some clusters with 20 billion documents to serve 300,000 customers and 2 million monthly… | ||
| The Aryn Journey - Growing a Gen AI Startup on OpenSearch and the Road Ahead | Mehul Shah | 24m |
Aryn builds a GenAI stack on OpenSearch, adding a RAG search processor and conversational memory to enable retrieval-augmented conversational search; the… | ||
| Welcome Back | Kassian Wren | 02m |
Announces the event schedule and Wi Fi via a QR code on your badge, lists sponsors by tier — gold: Atlassian, AWS, Uber, Yugabyte DB; silver: Data Stacks… | ||
| Beyond Vector Search: Building Ultra-Resilient GenAI Applications... | Karthik Ranganathan | 15m |
Argues that standalone vector stores create operational friction around data synchronization, ACID guarantees and enterprise resilience and demonstrates a… | ||
| Scaling OpenSearch: Next-Generation Shard Allo... | Rishab Nahata, Arpit Bandejiya, Himshikha Gupta | 39m |
Describes concrete changes that cut full-cluster reroute from 232 minutes to about 40 minutes on a 1,000-node test with 500k shards on 1,000 nodes… | ||
| CYOP: Create Your Own Plugin | Gaurav Gupta, Nishtha Mittal, Craig Perkins | 32m |
Step‑by‑step plugin build from scaffold to cluster: shows creating plugin zip with plugin descriptor.properties and optional security.policy, building with… | ||
| 10 Billion Downloads: Insights and Trends in Open Source | Avi Press | 30m |
Dataset covers over a trillion downloads (about 2 billion/day) from registries and SDK hooks like scarf.js postinstall telemetry, with ~100 million anonymized… | ||
| Migrating Queries From Lens and Saved Objects -- Data Tangles To Data... | Sarah Hudspeth, Jason Ng | 26m |
Chronosphere engineers extract nested ndjson saved objects and parse KQL/DQL/Lucene shards into reproducible queries, using Jupyter notebooks to inspect… | ||
| Resilient Segment Replication With Adaptive Refresh | Vigya Sharma | 40m |
Segment replication causes huge checkpoints that trigger major page faults; implementing byte-sized commits (multiple Lucene commit points per replication… | ||
| Leveraging Digital Twins and OpenSearch for Personalized Customer Experiences | Satej Sahu | 42m |
Builds a customer digital twin by generating PySim simulated shoppers and streaming synthetic and real events to S3 where Lambda pseudonymizes data and indexes… | ||
| Enhancing Query Performance Through Intelligent Data Co-location | Rishav Sagar, Tejas Shah | 43m |
Claims a segment creation and merge strategy for OpenSearch that co-locates related documents using a user-defined grouping field or script in index mapping… | ||
| Optimizing LLM Performance With Caching Strategies in OpenSearch | Uri Rosenberg, Sherin Chandy | 36m |
Shows an OpenSearch-backed LLM cache that combines exact key and semantic matching, integrates with LongChain GPT-cache patterns and OpenSearch vector stores… | ||
| From Bottlenecks To Breakthroughs: Optimizing Aggregation in OpenSearch for Lightning-Fas... | A. Jain | 29m |
Original monthly date-histogram that took ~7–8s drops to ~162ms when rewritten as per-month range queries and to ~2s when run serially, achieved by leveraging… | ||
| From LLM-as-a-Judge To Human-in-the-Loop: Rethinking Evaluat... | Eric Pugh, Fernando Rejon Barrera | 33m |
Adds LLM-as-a-judge support to OpenSearch 3.1 Search Relevance Workbench and ships a prototype RAGElo ELO-style framework that uses pairwise LLM comparisons to… | ||
| Beyond RAG - Going From Search to Analytics on Unstructured Data with Aryn | Mehul Shah | 43m |
Builds an AI-powered unstructured data warehouse called Aryn that parses PDFs into images, tables, figures and text with a custom table extractor and VLMs… | ||
| Deep Dive Into OpenSearch Pull-Based Ingestion at Uber | Varun Bharadwaj, Yupeng Fu | 43m |
Uber contributes a pull-based ingestion engine into OpenSearch 3.0 that actively reads from Kafka and Kinesis via plugins, stores per-document offsets and a… | ||
| Operating an OpenSearch Powered E-commerce Search Engine at Scale | Lucian Precup, Amine Gani | 46m |
Operating an OpenSearch e‑commerce stack for 1000+ stores uses Adelean’s A2 platform with concrete ops rules: autocomplete uses at least four small indices and… | ||
| Ingestion at Scale: Open Source Pipelines for Getting Data Into Open... | Anurag Gupta, Kevin Fallis | 42m |
Shows concrete pipelines using Fluent Bit, Data Prepper, OpenTelemetry Collector and Kafka to feed OpenSearch with parquet output to S3 now supported in Fluent… | ||
| From Bug To PR: Turning Customer Pain Points Into Upstream Contributions | Brian Graf, Alex Bunday | 22m |
NetApp Instaclustr converts customer bug reports into upstream fixes by triaging issues, checking existing GitHub work, forking and coding fixes, deploying via… | ||
| Autoscaling Search and Analytics in Kubernetes: The OpenSearch Operator Advantage | Pawel Wlodarczyk | 36m |
A concrete autoscaling pipeline ties the OpenSearch Operator to GitOps and monitoring: OpenSearch Operator uses a smart scaler and pre‑drain to avoid red… | ||
| Why Content (no, Not Just Docs) Needs More From the Open-Source Community | Kassian Wren | 20m |
Argues that project context is as important as code and claims teams should budget content time — roughly 1.5 hours of content per hour of feature work —… | ||
| Building Modern Observability Architecture With OpenSearch, OS Ingest... | Luis Tiani, Lucas Vieira | 37m |
A working observability stack is assembled: AWS Distro for OpenTelemetry injects traces and generates metrics via a span-to-metric processor, traces and logs… | ||
| Query Understanding With Large Language Models: Techniques and... | Hajer Bouafif, Cédric Pelvet | 51m |
Shows an OpenSearch pipeline that uses LLMs (via Strands agents and Amazon Bedrock in the demo) to do query expansion and filter extraction, turning “black… | ||
| Mastering OpenSearch Benchmark: Advanced Techniques for Power Users | Govind Kamat, Rishabh Singh | 39m |
Shows how OpenSearch Benchmark actually gets extended: register a Python generate_query_with_random_status parameter source to emit queries that pick HTTP… | ||
| Kubernetes Under the Magnifying Glass: Observability With OpenSearch | Arnab Chatterjee | 21m |
Shows how Kubernetes logs, metrics and traces land in a single OpenSearch backend using Fluent Bit daemonsets for logs and an OpenTelemetry collector for… | ||
| Growing a Vibrant Open Source OpenSearch Community in China | Charlie Yang | 28m |
China grows OpenSearch from near-zero to a top contributing region: Chinese repos now lead GitHub star counts for OpenSearch, China ranks third for active core… | ||
| Managing the Managed: Advanced Automation and Security for Amazon... | Luciano Taranto, Derrike Nunn | 35m |
OpenSearch clusters, indices, saved objects, role mappings and Flow Framework workflows deploy from a single CDK command using the open source CDK ProServe lib… | ||
| Building Next-gen Search With OpenSearch | Pietro Mele, Lucian Precup | 36m |
OpenSearch 3.x exposes agentic RAG: register a model (connector + model group), create a plan-execute-reflect agent with conversation index memory, and bind… | ||
| Use OpenSearch as Your Vector Search Database To Optimize GenAI Accuracy! | Anil Inamdar | 28m |
Shows how OpenSearch becomes a production vector DB: enable the KNN plugin, store dense and sparse embeddings alongside inverted indexes, and use chunking… | ||
| Beyond Single Vectors: Multi-Vector Search for Enhanced... | Praveen Mohan Prasad, Gene Alpert | 52m |
Multi-vector late-interaction (ColBERT / ColPaLi) keeps token/patch vectors and uses max-sim token→token scoring, implemented in OpenSearch via nested vector… | ||
| A Deep Dive Into OpenSearch Plugin Architecture and Introducing Hot... | Pranav Garg, Aman Kumar | 37m |
A prototype implements load/register/remove APIs to add plugin classes to the JVM, update mutable registries (analysis Registry changes from immutable maps to… | ||
| Agentic AI Meets Search: Real-World Use Cases With... | Aditya Soni, Seema Saharan, Anshika Tiwari | 37m |
OpenSearch 3.0 experimental MCP enables agents to call search and observability data as typed tools and the demo wires a Streamlit UI to an MCP connector using… | ||
| Implementing Piped Processing Language in OpenSearch Via Apache Calcite | Lantao Jin, Heng Qian | 41m |
OpenSearch integrates Apache Calcite into a new core engine v3 to compile PPL into Calcite RelNodes, reuse v2 scan logic, and generate physical plans with… | ||
| Mastering OpenSearch Snapshots: From Basics To Advanced Strategies | Ashish Singh, Chaitanya KSR | 48m |
Covers concrete ops and numbers: repository tests on S3/Azure/GCS, hashed prefix snapshots to spread backend partitions, and shallow v2 snapshots that reduce… | ||
| OpenSearch Joins a Foundation: A Blueprint for Community-Led Migration | Jiaxiang Zhu, Saurabh Singh | 36m |
OpenSearch transfers 120+ repos, domains and GitHub org ownership to the Linux Foundation, creates a Technical Steering Committee and SIGs (notably a Big Build… | ||
| Extending OpenSearch: A Practical Guide To Building Custom Plugins | Lucian Precup, Oceane Piedallu | 24m |
Six concrete plugins with code on GitHub: a smart ASCIIfolding rules analyzer that preserves accents for specified words (fixes French edge cases), an ingest… | ||
| Derived Source: Slash Storage Costs Without Losing Data in Open... | Mohit Godwani, Tanik Pansuriya | 37m |
Derived Source reconstructs the _source at fetch time from doc values and stored fields and is configurable via an index setting. It ships in OpenSearch 3.2… | ||
| Digging Into OpenSearch’s Memory: Lucene and JVM Garbage Collection Under the... | Yeonghyeon Ko | 39m |
Indexing in OpenSearch generates a storm of short‑lived Lucene objects that drive frequent young GCs, high promotion rates and wasted heap; attaching visual VM… | ||
| From Chaos To Clarity: Centralizing Observability With OpenSearch at Scale | Ido Ziv | 35m |
Calura migrates observability to OpenSearch using AWS managed OpenSearch with Data Prepper for ingest, Fluent Bit + Prometheus exporters and OpenTelemetry for… | ||
| Orchestrating Multimodal RAG With Agentic Workflows | Bobby Mohammed, Surya Kari | 31m |
Built a multimodal RAG pipeline that combines OpenSearch MCP server, Quen2.5 VL (hosted on SageMaker AI) and AWS Strands as an orchestrator to let agents… | ||
| Innovating at Scale: Lessons from Pegasystems' OpenSearch Journey | Amanda Katona, Satya Mishra | 35m |
Pegasystems migrates from embedded Elasticsearch to managed OpenSearch with NetApp Instaclustr, using Kafka and Cassandra as backing services, running in… | ||
| AI-Ready Data Foundation Using Amazon OpenSearch | Navnit Shukla, Gagan Brahmi | 38m |
Shows a working OpenSearch pipeline that ingests PDFs, emails and logs, creates vector-enabled indexes, and serves hybrid semantic plus keyword queries with… | ||
| Accelerating OpenSearch With Streaming: Apache Arrow, Flight, Data... | Saurabh Singh, Harsha Vamsi | 45m |
A new streaming framework integrates Apache Arrow with Flight and gRPC and a DataFusion-capable query layer to stream per-segment Arrow batches to the… | ||
| AI & OpenSearch: Confronting the New Frontier of Security & Compl... | Hugo Huang, Mohamed Nsiri | 31m |
A survey of 500+ IT and security leaders finds 60% basic or no security for AI/ML systems and 43% are very or extremely concerned about securing the AI stack… | ||
| Securing Data With Cedar | Michael Schwartz | 30m |
Demo plugin embeds a Cedar-based Java PDP into OpenSearch to filter Search API results per-document, mapping claims from tokens to Cedar objects and emitting… | ||
| DTEX's OpenSearch Journey: From Migration to Innovation | Prashant Agrawal, Sulabh Jain | 40m |
Shows a live migration using the OpenSearch Migration Assistant and its reindex-from-snapshot flow that takes snapshots from Elasticsearch (supports 1.x… | ||
| From Kernel To Cloud: Building a Complete Open Source Observ... | Anirudha Jadhav, Shenoy Gurudatt | 31m |
Claims a full open source pipeline that captures kernel-level telemetry and surfaces root causes; uses Inspektor Gadget eBPF gadgets (CPU profile, OOM kill… | ||
| Operating OpenSearch at Scale: Lessons From Managing 10,000+ Clusters Across... | Hariharan Gandhi | 52m |
Run a K8s-based platform that manages 11,000 clusters and ~60,000 persistent volumes, uses custom Kubernetes operators plus per-namespace maintainers to… | ||
| Next-Gen Search: How Uber and the OpenSearch Community Built a Cloud Native 3.0 | Shubham Gupta | 31m |
Uber describes moving from a Lucene lambda pipeline to contributions that shape OpenSearch 3.0: a decoupled reader and writer roles pattern with writers… | ||
| Introducing OpenSearch’s Search Relevance Workbench! | Eric Pugh, Stavros Macrakis | 39m |
Search Relevance Workbench shows how to measure and improve search quality end to end: it introduces Search Configurations as named, versioned… | ||
| Work Smarter, Not Harder: Discover Your OpenSearch Clusters Breaking... | Mike Oviedo, Rishabh Singh | 23m |
OpenSearch Benchmark adds Redline Testing, a feedback-driven load test that starts at zero clients and dynamically ramps clients up or down by monitoring… | ||
| OTeL/Piped Processing Language: Simplifying Observability Queries With Open... | Peng Huo, Ashwin Pc | 37m |
OpenSearch Piped Processing Language (PPL) provides a Unix-style pipeline for observability queries, claiming Unix-like pipe syntax and compatibility with… | ||
| OpenTelemetry With OpenSearch: Use-Cases, Status and Roadmap | Itamar Syn Hershko | 29m |
Explains how to ingest OpenTelemetry traces, metrics and logs into OpenSearch using the OpenTelemetry Collector (recommend the Amazon distribution) and… | ||
| Integrating Realtime User Behavioral Embeddings Into OpenSearch | Trey Grainger | 41m |
Uses OpenSearch User Behavior Insights plus Spark notebooks to ingest clickstreams, train latent behavioral embeddings with alternating least squares, and… | ||
| Capacity Planning, and Scaling/Optimization for Vector Workloads | Jon Handler | 47m |
Capacity planning example targets 125 million products with ~2 KB metadata and 1536‑dim vectors. Exact storage hits ~3 TB for vectors (≈7 TB with replicas) and… | ||
| From Traces To Action: Auto-Instrumenting LLMs for Observability... | Aditya Soni, Anshika Tiwari | 32m |
Demonstrates auto-collecting traces, metrics, logs from an LLM app using OpenTelemetry auto-instrumentation, enriching requests with model, token and latency… | ||
| OpenSearch and the Path to a Trillion-scale Vector Database | Vamshi Vijay Nakkirtha | 43m |
OpenSearch adds concrete engineering: disk-optimized vector search uses binary quantization + HNSW to scan compressed bits, then rescoring on full-precision… | ||
| Advancing OpenSearch With gRPC and Protobuf at Uber: A High-Performance... | Karen Xu, Shuyi Zhang | 42m |
Uber adds a parallel gRPC + Protobuf transport to OpenSearch 3.2 that runs on separate ports and requires an enablement setting, with generated proto libraries… | ||
| Repeatable Benchmarking: An Exploration of OpenSearch Vs Elasticsearch | Evan Downing | 22m |
Compares OpenSearch v2.17 and Elasticsearch v8.15 using an open, automated framework running OpenSearch Benchmark on AWS via GitHub Actions, deploying fresh… | ||
| The Imitation Game: Encapsulate Your OSB Workload at Scale | Ian Hoang, Govind Kamat | 36m |
OpenSearch Benchmark now generates privacy-safe, production-like datasets and streams them into tests at terabyte scale. It offers synthetic data generation… | ||
| OpenSearch Flow and Agent Frameworks: Easy AI | Jon Handler | 43m |
OpenSearch 2.13 adds a flow framework and 3.0 ships workflow templates plus an agent framework that runs tool chains; the demo uses the neural plugin (ingest… | ||
| Vital Signs: OpenSearch and Bioobservability | Nate Boot | 32m |
A long‑stay EEG records a tonic‑clonic seizure and pins the misfiring to the left temporal lobe; seven out of ten epilepsy cases show no root cause. OpenSearch… | ||
| Beyond the Code: Evolving OpenSearch's Release Process for... | Jiaxiang Zhu, Sayali Gaikawad | 42m |
OpenSearch automates builds, tests, signing and publishing but keeps concrete controls: incremental builds cut build time ~34%, parallelized workflows shave… | ||
| OpenSearch Migrations: Modernize Without Worry | Peter Nied, Brian Presley | 42m |
Migration Assistant automates multi‑version upgrades by combining reindex from snapshot to backfill historical data and capture and replay (Kafka-backed) to… | ||
| GenAIOps - OpenSearch for AI & GenAI Platform Observability | Ramesh Kumar Manickam, Rama Pabolu | 20m |
End-to-end GenAIOps pipeline using OpenTelemetry -> OpenSearch -> Jaeger collects metrics, logs, traces and baggage from Kubernetes-hosted LLMs, Vector DBs… | ||
| Using OpenSearch as a Database With DataFusion | Marc Handalian, Michael Froh | 42m |
Integrates DataFusion with OpenSearch to run SQL-style planning, joins, and aggregations by implementing a DataFusion table provider that executes shard… | ||
| Introducing the OCSF Normalization Playground for Preparing Data for... | Kevin Fallis, Chris Helma | 42m |
Tool demonstrates rapid generation of OCSF 1.1.0 transformations using a GenAI assistant and produces runnable Python extraction/transformation code that… | ||
| Building Intelligent Search Systems With Agentic RAG and OpenSearch | Vivek Gautam, Rushabh Lokhande | 40m |
Demonstrates an Agentic RAG pipeline that uses OpenSearch vector DB with Amazon Titan embeddings, an agent built on Strands agent SDK, and a Lambda/API Gateway… | ||
| Unlocking GPU Powers for Vector Search: OpenSearch's New Frontier | Navneet Verma, Jay Deng | 47m |
Shows a drop‑in architecture that uploads segment vectors to an object store and offloads index building to a remote GPU build fleet, using NVIDIA CAGRA to… | ||
| Fine-Grained Authorization for Secure RAG With OpenSearch | Juan Pablo Noreña | 17m |
A running on‑prem RAG proof of concept uses OpenSearch as the vector DB, a router that builds kNN queries, and OpenFGA to assert per‑user permissions so only… | ||
| A Visual AI Search Flow Builder for OpenSearch: Build GenAI and Vector... | Tyler Ohlsen, Owais Kazi | 38m |
A low‑code UI plus a backend plugin actually provisions end‑to‑end OpenSearch ML stacks with a single API call: it registers connectors and models (Cohere… | ||
| Sharding 2.0: Unlocking Infinite Search Scalability in OpenSearch | Vikas Bansal, Paras Jain | 22m |
OpenSearch implements in-place physical shard splitting that replaces a hot root shard with multiple child shards using root-child shard model and mutually… | ||
| Vector Search Beyond the Hype: Real-World Implementation and Performance Insights | Samuel Herman | 44m |
Shows concrete outcomes from building the jVector KNN plugin for OpenSearch: inline vectors reduce random IO and often lower latency versus aggressive… | ||
| Flexible OpenSearch Data Management With Apache Iceberg: Data... | Sotaro Hikita, Shuhei Fukami | 39m |
Proposes using Apache Iceberg as the staging layer between sources and OpenSearch, demonstrating point‑in‑time loads, cost‑effective snapshot versioning and… | ||
| Keynote: Welcome Back | Eric Pugh | 02m |
Welcome back remarks thank sponsors listed as Lassian AWS Uber Yugabyte DB data stacks materialize NetApp Instacluster and Arian Elatra Greer Infino and Pulse… | ||
| Unconference: OpenSearch is not a database! | Gene Alpert | 09m |
Argument: OpenSearch lacks no integrity enforcement and data independence, so schemas changing force app updates. Evidence: customers repeatedly hit join… | ||
| Unconference: Python Now! An OpenSearch world with/in/on Python | Eric Pugh, Charlie Yang | 14m |
Panel argues for bringing real Python into OpenSearch and outlines concrete paths: a JVM-embedded interpreter called grow pi for direct Java–Python access… | ||
| Unconference: Explainable Vector Search | Praveen Mohad Prasad | 15m |
Claim: single-vector underperforms BM25 on complex logical queries, supported by a 50k document dataset with 1k queries where single-vector embedder trails… | ||
| Unconference: How will AI + Search (IR) converge? | Stavros Macrakis | 09m |
Argues that information retrieval has a 50 year history that AI folks often ignore and that popular fixes like vector DB vs search misunderstand the need for… | ||
| Unconference: Self-optimizing deep research agents | Jakub Zavrel | 14m |
Claims that modular, non-differentiable deep research agents can self-optimize using an LLM to critique and rewire prompts and pipelines, evaluated on… | ||
| Unconference: Context is all you need? | Navnit Shukla | 12m |
Claim: context selection — not bigger models — fixes hallucination and cost for production GenAI. Demonstration-level pattern: ingest core documents into… | ||
| Highlights — San Jose, California (Sept 8–10) | 01m | |
Three days of clips compress keynote claims, three technical deep dives, community panels and live demos showing working OpenSearch deployments and migrations… | ||
| nothing matches | ||