Discover why Altinity.Cloud is the preferred managed service for ClickHouse®

Altinity.Cloud® and ClickHouse Cloud are both cloud platforms for managed ClickHouse. Learn why enterprises choose Altinity.Cloud over other managed ClickHouse vendors.

Get started instantly. No credit card required.

Altinity.Cloud vs ClickHouse Cloud FAQ

Altinity.Cloud runs ClickHouse on Kubernetes using the Altinity Kubernetes Operator for ClickHouse. Each tenant environment is a dedicated Kubernetes with complete isolation from other tenants. You can scale nodes up or down. You can increase storage at any time. Altinity.Cloud can manage clusters in any Kubernetes cluster using Altinity.Cloud in your VPC (originally called Altinity.Cloud Anywhere).

ClickHouse Cloud uses a serverless model with decoupled compute and storage that is similar to BigQuery. It stores data on object storage and transparently spins up server nodes to handle query traffic. It depends on local caching to help speed access to data when querying. Tenants share the underlying cloud infrastructure, though there is also a dedicated option. 

Altinity.Cloud runs in any AWS, GCP, and Azure region. You can also use Altinity.Cloud to operate in additional clouds like Digital Ocean, and others. Contact us for more information. 

ClickHouse Cloud runs in a limited number of AWS and GCP regions. Azure and other clouds are not supported yet.

Altinity.Cloud bills for allocated compute and storage, regardless of the level of use. Altinity.Cloud in your VPC (Altinity.Cloud Anywhere) charges only for compute.The pricing model in both cases is simple and predictable. You can get estimates using our pricing calculator. We offer discounts for volume usage and prepayment. We support multiple ways to reduce spend substantially, such as turning off unused ClickHouse servers. 

ClickHouse Cloud pricing is similar to BigQuery. It charges for object storage to hold data plus the cost of compute to execute queries. Dev and production deployment use a pure serverless model that is like paying “by the sip” for drinks. There is also a dedicated model that allows customers to reserve capacity. See the ClickHouse Cloud pricing page for details.

It depends on your use case. Serverless data services based on object storage – like ClickHouse Cloud – are inexpensive if you query only a small fraction of your data regularly. But if your usage suddenly increases, so does your bill. Also, many users find “by the sip” billing confusing. Changing table sort order or tweaking SQL queries can result in unpredictable cost increases. 

The Altinity.Cloud model allows users to read or write any amount of data at the same fixed cost. You can adjust costs quickly and predictably by rescaling cluster resources. Developers can change applications to offer new features without worrying that they will incur additional charges. It’s a good model for SaaS providers.

It depends on the use case. ClickHouse Cloud has a serverless model of operation and uses object storage as a backing store for data. It is fast when data is in cache but can be slower when large amounts of data must be fetched from object storage. Performance in non-dedicated accounts may be slower due to noisy neighbors. 

Altinity.Cloud supports a wide range of VM types with capacity from 2 to 64 vCPUs. There are no limitations on the number of shards or replicas. Users can configure clusters to handle a wide range of high-performance workloads. Altinity.Cloud accounts run on dedicated Kubernetes clusters so there are no “noisy neighbor” problems. Altinity support engineers will help you tune ClickHouse to achieve maximum performance. 

Altinity.Cloud in your VPC (Altinity.Cloud Anywhere) lets users run ClickHouse clusters in their own cloud account and VPCs. You register a Kubernetes cluster and Altinity.Cloud manages ClickHouse there. ClickHouse code and data are local. Users have complete control over security, data, and resource costs. 

ClickHouse Cloud runs only in its own account using its own VPCs.

Altinity.Cloud runs the same open source ClickHouse code you run on your laptop, including Altinity Stable Builds. If it works on your laptop, it will work on Altinity.Cloud. The few exceptions are features related to networking such as interfaces for gRPC, MySQL, and other protocols. We can also enable these if you need them. 

ClickHouse.Cloud is generally compatible with open source ClickHouse but has some code differences to support cloud operation. It does not support certain storage engines, including Kafka, limits dictionaries, and applies other restrictions like disabling experimental features. Read the compatibility guide for an up-to-date list.

Altinity supports ClickHouse wherever you need it to run. Since Altinity.Cloud ClickHouse versions are 100% open source, you can migrate to self-managed operation or even another cloud vendor without application changes. With Altinity.Cloud Anywhere or Altinity Enterprise support, there’s not even a “migration.” ClickHouse software and data remain in your possession at all times even if you unsubscribe from Altinity. 

ClickHouse Cloud does not support operation of ClickHouse outside their cloud. If you unsubscribe from ClickHouse Cloud, your data is no longer available. To migrate away from ClickHouse Cloud you must move the data to another managed service or to your own servers.You may also find that features differ when running pure open source ClickHouse.

Try them out! Both platforms offer free trials. Here are links for the Altinity.Cloud signup page and the ClickHouse Cloud signup page. Tell us what you think!

Running ClickHouse on Altinity.Cloud cut our costs by 25% and boosted our performance. Even better, we can also store data with higher granularity with ClickHouse–down to mouse click moves–than we could have done with Elasticsearch. We can show aggregate results to users and let them drill down into detailed source data for deeper, more insightful analysis than was previously possible.

Software engineer, Product Analytics Company

Speed up your ClickHouse queries by 50%. Watch the 60 min tutorial.

  • Learn the basic framework for performance (I/O, compute, ClickHouse structures like compression and indexes, etc).
  • Discover how to evaluate performance (ClickHouse system tables, EXPLAIN, etc)
  • Improve performance for common query use cases ranging from MergeTree data on block storage to Parquet files in data lakes.