AWS announces AWS Interconnect – multicloud (preview), providing simple, resilient, high-speed private connections to other cloud service providers. AWS Interconnect - multicloud is easy to configure and provides high-speed, resilient connectivity with dedicated bandwidth, enabling customers to interconnect AWS networking services such as AWS Transit Gateway, AWS Cloud WAN, and Amazon VPC to other cloud service providers with ease.
OpenSearch Search/Indexing Latency Graphics empty on AWS Console after adding Coordinator nodes in cluster
we had a large cluster 3 master nodes and more than 40 data nodes. To improve query performance, we introduce the coordinator nodes as 3 nodes. But on AWS dashboard, which the screenshot attached, search latency and indexing latency graphics broken and shows empty. Is it expected behavior of AWS Console?
- Newest
- Most votes
- Most comments
Yes, this is expected behavior in the AWS Console after adding dedicated coordinator nodes to your OpenSearch cluster. When you add coordinator nodes to your cluster, the search and indexing requests are now being coordinated through these dedicated nodes rather than directly through the data nodes.
In an OpenSearch cluster with dedicated coordinator nodes, the coordination responsibilities for both search and indexing operations are offloaded from data nodes to these specialized nodes. The coordinator nodes handle tasks like splitting bulk indexing requests into shard-specific subrequests, routing them to corresponding shards, querying various shards for search requests, and performing additional computation for queries with aggregations.
The OpenSearch Dashboards process also typically runs on coordinator nodes rather than data nodes in clusters that have dedicated coordinator nodes. This separation of responsibilities means that the metrics for search and indexing latency may now be associated with the coordinator nodes instead of appearing in the data node metrics that you're viewing.
To get a complete picture of your cluster's performance after adding coordinator nodes, you should check the metrics for both your data nodes and coordinator nodes. The coordinator nodes will now be handling the initial processing of search and indexing requests before distributing the work to data nodes, which explains why these particular graphs may appear empty in the view you're looking at.
Sources
Improve OpenSearch Service cluster resiliency and performance with dedicated coordinator nodes | AWS Big Data Blog
Restarting the OpenSearch Dashboards process on a node in Amazon OpenSearch Service - Amazon OpenSearch Service
Dedicated coordinator nodes in Amazon OpenSearch Service - Amazon OpenSearch Service
Relevant content
- asked 22 days ago
- AWS OFFICIALUpdated 4 months ago
So, could we say that this is a bug on AWS Console interface? Because in the coordinator node side, there is no graphic to show these information.