Managed Cluster

Amazon OpenSearch Service is a fully managed service that makes it easier to deploy, operate, and scale OpenSearch (now supporting Apache Lucene 9) clusters securely and cost-effectively on AWS. OpenSearch Service offers direct access to the Elasticsearch and OpenSearch APIs, which helps make your existing code and applications using Elasticsearch or OpenSearch work seamlessly with the service.

This fully managed service provisions all the resources for your OpenSearch cluster and launches it in any AWS Region of your choice within minutes. It automatically detects and replaces failed OpenSearch nodes, which reduces the overhead associated with self-managed infrastructures. You can scale your cluster horizontally or vertically—up to 3 petabytes of data—with zero downtime through a single API call or a few steps in the AWS Management Console. With this flexibility, OpenSearch Service can support any workload from single-node development clusters to production-scale, multinode clusters. 

Amazon OpenSearch Service Multi-AZ with Standby provides enhanced availability and consistent performance, making it suitable for production workloads. In case of infrastructure failures like node drop, instance failure, loss of AZ connectivity, or a single AZ failure, OpenSearch Service performs an automatic failover from active to standby nodes in under a minute so that you can continue to index and search data.

OpenSearch Service also provides a robust set of visualization tools and capabilities with support for OpenSearch Dashboards and Kibana (for Elasticsearch versions up to 7.10), without any additional fees. (OpenSearch Dashboards and Amazon Managed Grafana are both forks of Kibana).

This fully managed solution includes features such as fine-grained access control, alerting, Index State Management, and SQL support, among others. The OpenSearch Service feature set originates from the needs of customers like you and through open-source community initiatives like OpenSearch.

Additionally, OpenSearch brings together the power of traditional search, analytics, and vector search into one complete package when used as a vector database. The vector engine for Amazon OpenSearch Serverless introduces a simple and scalable vector storage and search capability to developers, giving them contextually relevant responses across billions of vectors in milliseconds by querying embeddings and text-based keywords in a single hybrid request. Developers can rely on the vector engine for a cost-efficient, secure serverless environment, allowing them to seamlessly transition from application prototyping to production.  

You must factor several considerations into your decision to move to a managed service, including preconfigured capabilities for operational stability, security, and performance of your search and analytics infrastructure.

This service includes support for multiple workloads, such as search, log analytics, distributed trace analytics and observability, security information and event monitoring (SIEM), and more of the following:

• Storage tiering with UltraWarm
  
• Fully compliant OpenSearch at no extra charge
• Large-scale OpenSearch operational expertise
  
• Built-in plugins that provide fine-grained access, vector-based similarity algorithms, or alerting and monitoring with the ability to automate incident response