Oracle Corporation
12/12/2024 | Press release | Distributed by Public on 12/12/2024 13:29
OCI Search with OpenSearch v2.15: Introducing New AI and ML Capabilities to Enhance your Application and Search
Today, we announce the general availability of OpenSearch version 2.15 in Oracle Cloud Infrastructure (OCI) Search with OpenSearch, along with several other significant improvements. With this update, OCI Search with OpenSearch supports the following key use cases:
AI and machine learning for applications
Semantic and conversational search capabilities
AI-powered capabilities in Observability
Security analytics
Since the launch in 2022 of our fully managed OpenSearch service in OCI as an alternative to Elasticsearch, customers have enthusiastically used its capabilities for in-app semantic search, log analytics, observability, and security monitoring. With version 2.11, we introduced features including semantic search, vector search, and retrieval-augmented generation (RAG), allowing search interactions to become more natural, whether for text or images.
Illustration of a multimodal search on OpenSearch on images
Figure 1: Multimodal search in OpenSearch
Now, let's explore together the features introduced by OpenSearch version 2.15. We've made significant improvements in the performance, efficiency, and resiliency, but more importantly, OpenSearch v2.15 elevates the potential of your search with groundbreaking AI and machine learning (ML) capabilities on OCI.
AI and ML innovations in OpenSearch 2.15
One of the major challenges of integrating AI and ML with OpenSearch has been the need to build custom middleware to connect ML models with the search engine. With OpenSearch 2.15, we're removing these barriers and opening a new range of use cases, specifically addressing AI and ML needs. You can now create AI integrations directly within OpenSearch with minimal effort, making advanced ML capabilities more accessible and powerful.
This release strengthens OCI Search with OpenSearch integration with other OCI services, such as OCI Data Science, OCI Generative AI and OCI Generative AI Agents, offering a cohesive and integrated solution for advanced AI workflows.
Key AI and ML features
The latest version offers the following AI and ML features:
Pretrained models integration: Use pretrained models for common ML tasks, such as text classification, sentiment analysis, and entity recognition, which were previously complex to implement.
OCI OpenSearch dashboard showing pretrained model registration
Figure 2: OCI OpenSearch dashboard showing pretrained model registration
RAG pipeline creation automation in OpenSearch: Easily create RAG pipelines to combine semantic search with generative AI, providing richer and context-aware responses in your search queries.
RAG pipeline creation simulation
Figure 3: RAG pipeline creation
External connector integration: Integrate external connectors with OCI Data Science models, train and tune them to generate embeddings, and improve your search accuracy, making it simpler to incorporate advanced ML capabilities within OpenSearch. Register and deploy the model of your choice with HuggingFace using AI Quick Actions in OCI Data Science.
Model registration utilizing OCI Data Science
Figure 4: Model registration utilizing OCI Data Science
LangChain integration for AI-powered search: OpenSearch's LangChain integration enables developers to seamlessly build generative chatbots and conversational search experiences by simplifying the use of large language models (LLMs) and vector search, eliminating the need for custom middleware while enhancing AI-driven applications with features such as OpenSearch vector store and RAG templates.
These AI advancements make it possible for you to create smarter, more responsive search applications, whether for enhanced search relevancy, anomaly detection in logs, or rich AI-powered interactions.
Other features added to OCI Search with OpenSearch
We're also releasing the following features in OCI Search with OpenSearch:
Cross-cluster search: Perform seamless searches across multiple OpenSearch clusters, allowing for scalability and a distributed environment without the complexity of managing independent clusters.
Prometheus plugin for observability: The new Prometheus plugin integrates directly with OpenSearch, allowing you to monitor metrics, visualize performance, and get real-time insights for better observability of search applications.
Searchable snapshots: A searchable snapshot index reads data from a snapshot repository on demand in real time at search time, rather than downloading all index data to cluster storage at restore time.
Data Prepper for log ingestion in limited availability: Data Prepper is now available in early access for preprocessing and transforming logs before they're ingested into OpenSearch, making it easier to prepare data for analysis and ensuring that ingested data is optimized for search and observability.
Try it yourself!
With these additional features, OCI Search with OpenSearch continues to support large-scale and distributed environments while providing the tools needed for effective monitoring and observability. This upgrade comes at no extra charge. Oracle Cloud Infrastructure maintains simple, transparent pricing, based on the number of nodes rather than the number of cores, unlike other hyperscalers.
Ready to explore the new features? Head over to the Oracle Cloud Console to start upgrading today. To learn more about how version 2.15 can benefit your organization, check out our documentation.
For more information, see the following resources:
Search with OpenSearch documentation
OpenSearch hands-on labs
Read customer testimonials from Prophecy and NetSuite
Oracle Architecture Center