We ran OpenSearch Benchmark against a three-node ClusterNest Standard cluster and the matching Amazon OpenSearch Service configuration, from a test machine inside AWS's own region. We measured bulk indexing, queries and both together, from a single client up to 64 in parallel, and priced each cluster at its list price.
Read itOpenSearch 3.9 is live on ClusterNest, with a native neural sparse search engine, leaner vector storage, and new observability tools in Dashboards.
Read itMost Kubernetes log pipelines run on a password: one credential for the log store, copied to every node, shared across clusters and almost never rotated. Kubernetes already gives every pod a signed identity that expires in minutes, and OpenSearch already knows how to check one. We joined the two and ended up with a logging pipeline that has no password anywhere in it, works across a fleet of clusters, and lets each cluster write only its own logs.
Read itManaged OpenSearch now trusts several identity providers at once. Employees, contractors, the company you just acquired and your CI pipelines all sign in through the system that already governs them, and the secrets behind it never leave the vault. SSO is still free on every tier.
Read itOpenSearch 3.8 is live on ClusterNest, with faster vector search, a broader set of AI agent integrations, and new tools for exploring logs without leaving Dashboards.
Read itOpenSearch 3.7 just landed on ClusterNest, with a strong focus on observability and query safety.
Read itWe sent 494,175 numbered log lines through Fluent Bit and Vector into OpenSearch and counted every one that arrived. With stock settings, one shipper lost 62 percent of a busy pod's logs and the other lost 85 percent, both without logging a single error. Five defaults were behind it, and each one is a one-line fix.
Read itBig milestone: OpenSearch 3.6 is live on ClusterNest, and it's the project's first Long-Term Support release, with 18+ months of guaranteed support, so it's a solid pick for clusters you don't want to babysit through every minor upgrade.
Read itWe shipped exactly 100 MB of logs into OpenSearch, once with Fluent Bit and once with Vector, and measured every byte that landed on disk. One copy came to about 1.5 times the raw size, the replica doubled that, and the search index, not the logs, turned out to be the biggest part of the bill.
Read itOpenSearch 3.3 and 3.4 are selectable on ClusterNest clusters. In the same change, OpenSearch 2.x is deprecated here: still running, no longer offered on new clusters.
Read itOpenSearch 3.2 can be selected on new and existing clusters. The console handles it, and Terraform provider 1.1.2 adds it to the version list.
Read itOpenSearch clusters can now snapshot to an S3 bucket you control, on AWS or any S3-compatible store. Authentication is an IAM role or static keys, and the copy is yours whatever happens to your account with us.
Read itOpenSearch 3.0 and 3.1 can now be selected on ClusterNest clusters. It is the first major release since 2.0, rebased on Lucene 10 and requiring Java 21, so it is a version worth reading the notes for before you move.
Read itAdvanced tier OpenSearch clusters are open to everyone. Eight vCPU, 64 GiB and 2 TB of NVMe per node, launched from the console with no quota request and no waiting for an approval.
Read itOpenSearch clusters can now delegate authentication to your identity provider over SAML or OIDC, so access follows the directory you already run. It costs nothing extra, on every tier.
Read itClusters can now be declared in Terraform rather than clicked into existence. The ClusterNest provider is out of preview: create, update and destroy OpenSearch clusters from the same repository as the rest of your infrastructure.
Read itTwo years after launching managed OpenSearch, we are closing the Elasticsearch service to new clusters. Existing customers keep running, and we will come to you about the migration rather than the other way round.
Read itIn 2013 you set up centralised logging by installing Elasticsearch, Logstash and Kibana, and the reason that was a good decision is that it was not really a decision. Ten years on, every letter has multiple credible answers, the licence under the first one has moved, and the integration work is yours. Here is what happened, what it costs, and the four questions I now use to pick a stack.
Read itThe service we said we were building in July is open. Managed OpenSearch clusters, priced by tier and node count like our Elasticsearch service, with OpenSearch Dashboards included and a migration we run with you.
Read itOpenSearch has reached a stable release. We said in January that the answer to the Elasticsearch relicensing was an engine that is genuinely open with more than one vendor behind it, and this is the one. A managed service is in build.
Read it