OpenSearch Service recently introduced support for Multi-AZ with Standby, a new deployment option for the search and analytics engine that provides 99.99% availability and better performance for business-critical workloads.
With Multi-AZ with Standby, OpenSearch Service reserves nodes in one of the AZs as standby making deployments resilient to potential infrastructure failures and simplifying configuration and management. Prashant Agrawal, senior search specialist solutions architect, and Rohin Bhargava, senior product manager, explain the advantages of the new option:
When an issue arises, such as a node becoming unresponsive, OpenSearch Service recovers by recreating the missing shards (data), causing a potentially large movement of data in the domain. This data movement increases resource usage on the cluster, which can impact performance. If the cluster is not sized properly, it can experience degraded availability, which defeats the purpose of provisioning the cluster across three Availability Zones.
According to the cloud provider, the new configuration option improves availability to 99.99% and ensures that domains follow recommended best practices, simplifying configuration and management.
OpenSearch Service is a managed option to deploy OpenSearch clusters, supporting OpenSearch and legacy Elasticsearch OSS, up to Elasticsearch 7.10. Multi-AZ with Standby requires a domain running on OpenSearch 1.3 or above, deployed across three availability zones with three (or a multiple of three) data nodes. Furthermore, only GP3- or SSD-backed instances and a subset of instance types are currently supported. While the service distributes the nodes and data copies across three AZs, Agrawal and Bhargava warn:
During normal operations, the standby nodes don’t receive any search requests. The two active Availability Zones respond to all search requests. However, data is replicated to these standby nodes to ensure you have a full copy of the data in each Availability Zone at all times.
OpenSearch Service still supports multi-AZ without standby, offering 99.9% availability and lower costs as all the cluster nodes can serve read requests.
The new capability of the successor of ElasticSearch Service rotates the standby AZ every 30 minutes to ensure the system is running and ready to respond to changes, with the AZ Rotation Metrics exposing the status of the cluster, showing active reads and active writes.
The Multi-AZ with Standby is not the only feature recently introduced for the managed search and analytics engine: AWS recently announced the GA of the controversial OpenSearch Serverless option, OpenSearch Ingestion, a serverless data collector, and Security Analytics. The new failover option is currently available in most but not all AWS regions where OpenSearch Service is supported.