A true alternative to Elastic Cloud for Elasticsearch ops automation

The latest release of ClusterControl —v1.9.3—provides support for Elasticsearch, the most popular full-text search and analytics engine around today. By supporting it, we’ve re-opened the door for Elasticsearch users searching for a way to fully automate end-to-end operations.

In this post, we’ll cover the current challenges that Elasticsearch users face and how ClusterControl’s growing support for this technology will change the way users can implement and manage their Elasticsearch databases.

Why Elasticsearch?

Introduced in 2010, Elasticsearch remains the most popular open-source search and analytics engine. In fact, it still maintains a top 10 spot in DB-Engines despite recent license controversy (covered below). Many users flock to Elasticsearch for its ability to provide log analytics, full-text search, business analytics, security analytics, and data visualizations at breakneck speeds.

To put it simply, Elasticsearch will likely keep its edge for the foreseeable future because it is distributable, easily scalable, and extremely fast, making it invaluable for those building modern applications that rely on complex search features, but it’s not without controversy.

Elasticsearch licensing changes have posed difficulties 

Elasticsearch users have run into some obstacles when it comes to options for managing the database after Elastic changed its license in 2021 to a dual licensing structure — Server Side Public License (SSPL) and the Elastic License — to protect its interests from hyperscalers. 

Effectively, this move to a “proprietary-lite” structure not only prevents cloud providers, but traditional DBaaS and other service providers like Aiven and Logz.io from offering managed ELK services. In a countermove, AWS and a number of partners created a fork, OpenSearch.

So now Elasticsearch customers have been left with three options:

1. Self-manage — This is fine if you’ve got the in-house expertise and resources to manage Elasticsearch ops; however, many users don’t and would rather implement full-lifecycle automation atop their deployments due to use cases that indicate large footprints.
2. Elastic Cloud — Pay for Elastic’s managed service and introduce vendor lock-in risks.
3. OpenSearch — Shift your workloads to OpenSearch which can also be a risky move due to its uncertain future, e.g. will Amazon Web Services and its small consortium of committers build enough steam behind it so that it gets maintained and developed.

All three of these options present unavoidable headaches and risks for Elasticsearch users, which is exactly why we decided to add support for Elasticsearch to our database catalog.

Why use ClusterControl for Elasticsearch?

ClusterControl’s support for Elasticsearch means there is still an opportunity for full-lifecycle automation without using Elastic’s own managed service or moving to OpenSearch. With ClusterControl, Elasticsearch 7.x and 8.x users of the open-source edition can now reliably automate database operations on-prem, in the cloud, or in hybrid environments.

So why should Elasticsearch customers consider using ClusterControl to manage its ops?

1. Using ClusterControl is the only way to get out-of-the-box, full-lifecycle automation support for Elasticsearch operations without going to Elastic’s managed service.

That’s because ClusterControl is a database automation platform that you control and not a traditional DBaaS managed by a third party. Result? You get the benefits of both—full access to open-source licenses and a solution to easily deploy, manage and monitor your HA Elasticsearch from a single control plane—without facing license risks.

2. ClusterControl enables customers to deploy environment-agnostic implementations of Elasticsearch.

With ClusterControl, you avoid environment lock-in, a common risk associated with traditional managed services that are either public cloud-based or provide alternative environments tied to a cloud provider’s ecosystem. Instead, you’re free to deploy anywhere you want, whether on-prem, in private and public clouds, or in a hybrid setup.

3. ClusterControl is ideal for search implementations, e.g. modern web apps, that don’t require the paid license features of Elasticsearch, yet require large implementations.

Elasticsearch implementations that require a large, distributed footprint, such as search, can be challenging to operationalize without a dedicated team. ClusterControl simplifies the process by providing one tool to manage the complete stack of database operations.

Ultimately, it boils down to the fact that ClusterControl provides the automation suite of a traditional DBaaS needed to confidently run large-scale deployments of Elasticsearch in any environment without license restrictions. Now, let’s check out ClusterControl’s initial features.

Current ClusterControl+Elasticsearch features

ClusterControl’s current support of Elasticsearch includes the following:

• Available via CLI or ClusterControl v2 UI (find out how to set it up here)
• Deploy one node for test or deployment environments
• Deploy three or more nodes for clustered deployments with master/data roles
• Basic User Authentication with username and password
• TLS/SSL API endpoint encryption
• Backup Management with local file storage repository

For full details on ClusterControl’s 1.9.3 release, visit the changelog.

Wrapping Up

We will continue rolling out more features for Elasticsearch in the upcoming ClusterControl releases. If you’re looking for a way to deploy, manage, and scale your Elasticsearch implementation, you can evaluate ClusterControl free for 30 days —no CC required.

In the meantime, we will be publishing additional blog posts on Elasticsearch use cases and how to implement them, installing Elasticsearch with ClusterControl, and tackling the rest of the ELK Stack (Elasticsearch, Logstash, and Kibana). So make sure you stay tuned for more updates by following us on Twitter and LinkedIn and subscribing to our newsletter. Talk soon!