Skip to main content
Skip to main content
Edit this page

Cloud changelog - 2024

December 20, 2024​

Marketplace subscription organization attachment​

You can now attach your new marketplace subscription to an existing ClickHouse Cloud organization. Once you finish subscribing to the marketplace and redirect to ClickHouse Cloud, you can connect an existing organization created in the past to the new marketplace subscription. From this point, your resources in the organization will be billed via the marketplace.

Force OpenAPI key expiration​

It is now possible to restrict the expiry options of API keys so you don't create unexpired OpenAPI keys. Please contact the ClickHouse Cloud Support team to enable these restrictions for your organization.

Custom emails for notifications​

Org Admins can now add more email addresses to a specific notification as additional recipients. This is useful in case you want to send notifications to an alias or to other users within your organization who might not be users of ClickHouse Cloud. To configure this, go to the Notification Settings from the cloud console and edit the email addresses that you want to receive the email notifications.

December 6, 2024​

BYOC (beta)​

Bring Your Own Cloud for AWS is now available in Beta. This deployment model allows you to deploy and run ClickHouse Cloud in your own AWS account. We support deployments in 11+ AWS regions, with more coming soon. Please contact support for access. Note that this deployment is reserved for large-scale deployments.

Postgres Change Data Capture (CDC) connector in ClickPipes​

This turnkey integration enables customers to replicate their Postgres databases to ClickHouse Cloud in just a few clicks and leverage ClickHouse for blazing-fast analytics. You can use this connector for both continuous replication and one-time migrations from Postgres.

Dashboards (beta)​

This week, we're excited to announce the Beta launch of Dashboards in ClickHouse Cloud. With Dashboards, you can turn saved queries into visualizations, organize visualizations onto dashboards, and interact with dashboards using query parameters. To get started, follow the dashboards documentation.

Query API endpoints (GA)​

We are excited to announce the GA release of Query API Endpoints in ClickHouse Cloud. Query API Endpoints allow you to spin up RESTful API endpoints for saved queries in just a couple of clicks and begin consuming data in your application without wrangling language clients or authentication complexity. Since the initial launch, we have shipped a number of improvements, including:

  • Reducing endpoint latency, especially for cold-starts
  • Increased endpoint RBAC controls
  • Configurable CORS-allowed domains
  • Result streaming
  • Support for all ClickHouse-compatible output formats

In addition to these improvements, we are excited to announce generic query API endpoints that, leveraging our existing framework, allow you to execute arbitrary SQL queries against your ClickHouse Cloud services. Generic endpoints can be enabled and configured from the service settings page.

To get started, follow the Query API Endpoints documentation.

Native JSON support (Beta)​

We are launching Beta for our native JSON support in ClickHouse Cloud. To get started, please get in touch with support to enable your cloud service.

Vector search using vector similarity indexes (early access)​

We are announcing vector similarity indexes for approximate vector search in early access.

ClickHouse already offers robust support for vector-based use cases, with a wide range of [distance functions]https://clickhouse.com/blog/reinvent-2024-product-announcements#vector-search-using-vector-similarity-indexes-early-access) and the ability to perform linear scans. In addition, more recently, we added an experimental approximate vector search approach powered by the usearch library and the Hierarchical Navigable Small Worlds (HNSW) approximate nearest neighbor search algorithm.

To get started, please sign up for the early access waitlist.

ClickHouse-connect (Python) and ClickHouse Kafka Connect users​

Notification emails went out to customers who had experienced issues where the clients could encounter a MEMORY_LIMIT_EXCEEDED exception.

Please upgrade to:

  • Kafka-Connect: > 1.2.5
  • ClickHouse-Connect (Java): > 0.8.6

ClickPipes now supports cross-VPC resource access on AWS​

You can now grant uni-directional access to a specific data source like AWS MSK. With Cross-VPC resource access with AWS PrivateLink and VPC Lattice, you can share individual resources across VPC and account boundaries, or even from on-premise networks without compromising on privacy and security when going over a public network. To get started and set up a resource share, you can read the announcement post.

ClickPipes now supports IAM for AWS MSK​

You can now use IAM authentication to connect to an MSK broker with AWS MSK ClickPipes. To get started, review our documentation.

Maximum replica size for new services on AWS​

From now on, any new services created on AWS will allow a maximum available replica size of 236 GiB.

November 22, 2024​

Built-in advanced observability dashboard for ClickHouse Cloud​

Previously, the advanced observability dashboard that allows you to monitor ClickHouse server metrics and hardware resource utilization was only available in open-source ClickHouse. We are happy to announce that this feature is now available in the ClickHouse Cloud console.

This dashboard allows you to view queries based on the system.dashboards table in an all-in-one UI. Visit Monitoring > Service Health page to start using the advanced observability dashboard today.

AI-powered SQL autocomplete​

We've improved autocomplete significantly, allowing you to get in-line SQL completions as you write your queries with the new AI Copilot. This feature can be enabled by toggling the "Enable Inline Code Completion" setting for any ClickHouse Cloud service.

Animation showing the AI Copilot providing SQL autocompletion suggestions as a user types

New "billing" role​

You can now assign users in your organization to a new Billing role that allows them to view and manage billing information without giving them the ability to configure or manage services. Simply invite a new user or edit an existing user's role to assign the Billing role.

November 8, 2024​

Customer Notifications in ClickHouse Cloud​

ClickHouse Cloud now provides in-console and email notifications for several billing and scaling events. Customers can configure these notifications via the cloud console notification center to only appear on the UI, receive emails, or both. You can configure the category and severity of the notifications you receive at the service level.

In future, we will add notifications for other events, as well as additional ways to receive the notifications.

Please see the ClickHouse docs to learn more about how to enable notifications for your service.


October 4, 2024​

ClickHouse Cloud now offers HIPAA-ready services in Beta for GCP​

Customers looking for increased security for protected health information (PHI) can now onboard to ClickHouse Cloud in Google Cloud Platform (GCP). ClickHouse has implemented administrative, physical and technical safeguards prescribed by the HIPAA Security Rule and now has configurable security settings that can be implemented, depending on your specific use case and workload. For more information on available security settings, please review our Security features page.

Services are available in GCP us-central-1 to customers with the Dedicated service type and require a Business Associate Agreement (BAA). Contact sales or support to request access to this feature or join the wait list for additional GCP, AWS, and Azure regions.

Compute-compute separation is now in private preview for GCP and Azure​

We recently announced the Private Preview for Compute-Compute Separation for AWS. We're happy to announce that it is now available for GCP and Azure.

Compute-compute separation allows you to designate specific services as read-write or read-only services, allowing you to design the optimal compute configuration for your application to optimize cost and performance. Please read the docs for more details.

Self-service MFA recovery codes​

Customers using multi-factor authentication can now obtain recovery codes that can be used in the event of a lost phone or accidentally deleted token. Customers enrolling in MFA for the first time will be provided the code on set up. Customers with existing MFA can obtain a recovery code by removing their existing MFA token and adding a new one.

ClickPipes update: custom certificates, latency insights, and more.​

We're excited to share the latest updates for ClickPipes, the easiest way to ingest data into your ClickHouse service. These new features are designed to enhance your control over data ingestion and provide greater visibility into performance metrics.

Custom Authentication Certificates for Kafka

ClickPipes for Kafka now supports custom authentication certificates for Kafka brokers using SASL & public SSL/TLS. You can easily upload your own certificate in the SSL Certificate section during ClickPipe setup, ensuring a more secure connection to Kafka.

Introducing Latency Metrics for Kafka and Kinesis

Performance visibility is crucial. ClickPipes now features a latency graph, giving you insight into the time between message production (whether from a Kafka Topic or a Kinesis Stream) to ingestion in ClickHouse Cloud. With this new metric, you can keep a closer eye on the performance of your data pipelines and optimize accordingly.


Scaling Controls for Kafka and Kinesis (Private Beta)

High throughput can demand extra resources to meet your data volume and latency needs. We're introducing horizontal scaling for ClickPipes, available directly through our cloud console. This feature is currently in private beta, allowing you to scale resources more effectively based on your requirements. Please contact support to join the beta.

Raw Message Ingestion for Kafka and Kinesis

It is now possible to ingest an entire Kafka or Kinesis message without parsing it. ClickPipes now offers support for a _raw_message virtual column, allowing users to map the full message into a single String column. This gives you the flexibility to work with raw data as needed.

August 29, 2024​

New Terraform provider version - v1.0.0​

Terraform allows you to control your ClickHouse Cloud services programmatically, then store your configuration as code. Our Terraform provider has almost 200,000 downloads and is now officially v1.0.0. This new version includes improvements such as better retry logic and a new resource to attach private endpoints to your ClickHouse Cloud service. You can download the Terraform provider here and view the full changelog here.

2024 SOC 2 Type II report and updated ISO 27001 certificate​

We are proud to announce the availability of our 2024 SOC 2 Type II report and updated ISO 27001 certificate, both of which include our recently launched services on Azure as well as continued coverage of services in AWS and GCP.

Our SOC 2 Type II demonstrates our ongoing commitment to achieving security, availability, processing integrity and confidentiality of the services we provide to ClickHouse users. For more information, check out SOC 2 - SOC for Service Organizations: Trust Services Criteria issued by the American Institute of Certified Public Accountants (AICPA) and What is ISO/IEC 27001 from the International Standards Organization (ISO).

Please also check out our Trust Center for security and compliance documents and reports.

August 15, 2024​

Compute-compute separation is now in Private Preview for AWS​

For existing ClickHouse Cloud services, replicas handle both reads and writes, and there is no way to configure a certain replica to handle only one kind of operation. We have an upcoming new feature called Compute-compute separation that allows you to designate specific services as read-write or read-only services, allowing you to design the optimal compute configuration for your application to optimize cost and performance.

Our new compute-compute separation feature enables you to create multiple compute node groups, each with its own endpoint, that are using the same object storage folder, and thus, with the same tables, views, etc. Read more about Compute-compute separation here. Please contact support if you would like access to this feature in Private Preview.

ClickPipes for S3 and GCS now in GA, Continuous mode support​

ClickPipes is the easiest way to ingest data into ClickHouse Cloud. We're happy to announce that ClickPipes for S3 and GCS is now Generally Available. ClickPipes supports both one-time batch ingest and "continuous mode". An ingest task will load all the files matched by a pattern from a specific remote bucket into the ClickHouse destination table. In "continuous mode", the ClickPipes job will run constantly, ingesting matching files that get added into the remote object storage bucket as they arrive. This will allow you to turn any object storage bucket into a fully fledged staging area for ingesting data into ClickHouse Cloud. Read more about ClickPipes in our documentation.

July 18, 2024​

Prometheus endpoint for metrics is now generally available​

In our last cloud changelog, we announced the Private Preview for exporting Prometheus metrics from ClickHouse Cloud. This feature allows you to use the ClickHouse Cloud API to get your metrics into tools like Grafana and Datadog for visualization. We're happy to announce that this feature is now Generally Available. Please see our docs to learn more about this feature.

Table inspector in Cloud console​

ClickHouse has commands like DESCRIBE that allow you to introspect your table to examine schema. These commands output to the console, but they are often not convenient to use as you need to combine several queries to retrieve all pertinent data about your tables and columns.

We recently launched a Table Inspector in the cloud console which allows you to retrieve important table and column information in the UI, without having to write SQL. You can try out the Table Inspector for your services by checking out the cloud console. It provides information about your schema, storage, compression, and more in one unified interface.

New Java Client API​

Our Java Client is one of the most popular clients that users use to connect to ClickHouse. We wanted to make it even easier and more intuitive to use, including a re-designed API and various performance optimizations. These changes will make it much easier to connect to ClickHouse from your Java applications. You can read more about how to use the updated Java Client in this blog post.

New Analyzer is enabled by default​

For the last couple of years, we've been working on a new analyzer for query analysis and optimization. This analyzer improves query performance and will allow us to make further optimizations, including faster and more efficient JOINs. Previously, it was required that new users enable this feature using the setting allow_experimental_analyzer. This improved analyzer is now available on new ClickHouse Cloud services by default.

Stay tuned for more improvements to the analyzer as we have many more optimizations planned.

June 28, 2024​

ClickHouse Cloud for Microsoft Azure is now generally available​

We first announced Microsoft Azure support in Beta this past May. In this latest cloud release, we're happy to announce that our Azure support is transitioning from Beta to Generally Available. ClickHouse Cloud is now available on all the three major cloud platforms: AWS, Google Cloud Platform, and now Microsoft Azure.

This release also includes support for subscriptions via the Microsoft Azure Marketplace. The service will initially be supported in the following regions:

  • United States: West US 3 (Arizona)
  • United States: East US 2 (Virginia)
  • Europe: Germany West Central (Frankfurt)

If you'd like any specific region to be supported, please contact us.

Query log insights​

Our new Query Insights UI in the Cloud console makes ClickHouse's built-in query log a lot easier to use. ClickHouse's system.query_log table is a key source of information for query optimization, debugging, and monitoring overall cluster health and performance. There's just one caveat: with 70+ fields and multiple records per query, interpreting the query log represents a steep learning curve. This initial version of query insights provides a blueprint for future work to simplify query debugging and optimization patterns. We'd love to hear your feedback as we continue to iterate on this feature, so please reach out—your input will be greatly appreciated.

Prometheus endpoint for metrics (private preview)​

Perhaps one of our most requested features: you can now export Prometheus metrics from ClickHouse Cloud to Grafana and Datadog for visualization. Prometheus provides an open-source solution to monitor ClickHouse and set up custom alerts. Access to Prometheus metrics for your ClickHouse Cloud service is available via the ClickHouse Cloud API. This feature is currently in Private Preview. Please reach out to the support team to enable this feature for your organization.

Other features​

  • Configurable backups to configure custom backup policies like frequency, retention, and schedule are now Generally Available.

June 13, 2024​

Configurable offsets for Kafka ClickPipes Connector (Beta)​

Until recently, whenever you set up a new Kafka Connector for ClickPipes, it always consumed data from the beginning of the Kafka topic. In this situation, it may not be flexible enough to fit specific use cases when you need to reprocess historical data, monitor new incoming data, or resume from a precise point.

ClickPipes for Kafka has added a new feature that enhances the flexibility and control over data consumption from Kafka topics. You can now configure the offset from which data is consumed.

The following options are available:

  • From the beginning: Start consuming data from the very beginning of the Kafka topic. This option is ideal if you need to reprocess all historical data.
  • From latest: Begin consuming data from the most recent offset. This is useful if you are only interested in new messages.
  • From a timestamp: Start consuming data from messages that were produced at or after a specific timestamp. This feature allows for more precise control, enabling you to resume processing from an exact point in time.

Enroll services to the Fast release channel​

The Fast release channel allows your services to receive updates ahead of the release schedule. Previously, this feature required assistance from the support team to enable. Now, you can use the ClickHouse Cloud console to enable this feature for your services directly. Simply navigate to Settings, and click Enroll in fast releases. Your service will now receive updates as soon as they are available.

Terraform support for horizontal scaling​

ClickHouse Cloud supports horizontal scaling, or the ability to add additional replicas of the same size to your services. Horizontal scaling improves performance and parallelization to support concurrent queries. Previously, adding more replicas required either using the ClickHouse Cloud console or the API. You can now use Terraform to add or remove replicas from your service, allowing you to programmatically scale your ClickHouse services as needed.

Please see the ClickHouse Terraform provider for more information.

May 30, 2024​

Share queries with your teammates​

When you write a SQL query, there's a good chance that other people on your team would also find that query useful. Previously, you'd have to send a query over Slack or email and there would be no way for a teammate to automatically receive updates for that query if you edit it.

We're happy to announce that you can now easily share queries via the ClickHouse Cloud console. From the query editor, you can share a query directly with your entire team or a specific team member. You can also specify whether they have read or write only access. Click on the Share button in the query editor to try out the new shared queries feature.

ClickHouse Cloud for Microsoft Azure is now in beta​

We've finally launched the ability to create ClickHouse Cloud services on Microsoft Azure. We already have many customers using ClickHouse Cloud on Azure in production as part of our Private Preview program. Now, anyone can create their own service on Azure. All of your favorite ClickHouse features that are supported on AWS and GCP will also work on Azure.

We expect to have ClickHouse Cloud for Azure ready for General Availability in the next few weeks. Read this blog post to learn more, or create your new service using Azure via the ClickHouse Cloud console.

Note: Development services for Azure are not supported at this time.

Our Private Link feature allows you to connect your ClickHouse Cloud services with internal services in your cloud provider account without having to direct traffic to the public internet, saving costs and enhancing security. Previously, this was difficult to set up and required using the ClickHouse Cloud API.

You can now configure private endpoints in just a few clicks directly from the ClickHouse Cloud console. Simply go to your service's Settings, go to the Security section and click Set up private endpoint.

May 17, 2024​

Ingest data from Amazon Kinesis using ClickPipes (beta)​

ClickPipes is an exclusive service provided by ClickHouse Cloud to ingest data without code. Amazon Kinesis is AWS's fully managed streaming service to ingest and store data streams for processing. We are thrilled to launch the ClickPipes beta for Amazon Kinesis, one of our most requested integrations. We're looking to add more integrations to ClickPipes, so please let us know which data source you'd like us to support. Read more about this feature here.

You can try the new Amazon Kinesis integration for ClickPipes in the cloud console:

Configurable backups (private preview)​

Backups are important for every database (no matter how reliable), and we've taken backups very seriously since day 1 of ClickHouse Cloud. This week, we launched Configurable Backups, which allows for much more flexibility for your service's backups. You can now control start time, retention, and frequency. This feature is available for Production and Dedicated services and is not available for Development services. As this feature is in private preview, please contact [email protected] to enable this for your service. Read more about configurable backups here.

Create APIs from your SQL queries (Beta)​

When you write a SQL query for ClickHouse, you still need to connect to ClickHouse via a driver to expose your query to your application. Now with our now Query Endpoints feature, you can execute SQL queries directly from an API without any configuration. You can specify the query endpoints to return JSON, CSV, or TSVs. Click the "Share" button in the cloud console to try this new feature with your queries. Read more about Query Endpoints here.

Official ClickHouse Certification is now available​

There are 12 free training modules in ClickHouse Develop training course. Prior to this week, there was no official way to prove your mastery in ClickHouse. We recently launched an official exam to become a ClickHouse Certified Developer. Completing this exam allows you to share with current and prospective employers your mastery in ClickHouse on topics including data ingestion, modeling, analysis, performance optimization, and more. You can take the exam here or read more about ClickHouse certification in this blog post.

April 25, 2024​

Load data from S3 and GCS using ClickPipes​

You may have noticed in our newly released cloud console that there's a new section called "Data sources". The "Data sources" page is powered by ClickPipes, a native ClickHouse Cloud feature which lets you easily insert data from a variety of sources into ClickHouse Cloud.

Our most recent ClickPipes update features the ability to directly upload data directly from Amazon S3 and Google Cloud Storage. While you can still use our built-in table functions, ClickPipes is a fully-managed service via our UI that will let you ingest data from S3 and GCS in just a few clicks. This feature is still in Private Preview, but you can try it out today via the cloud console.

Use Fivetran to load data from 500+ sources into ClickHouse Cloud​

ClickHouse can quickly query all of your large datasets, but of course, your data must first be inserted into ClickHouse. Thanks to Fivetran's comprehensive range of connectors, you can now quickly load data from over 500 sources. Whether you need to load data from Zendesk, Slack, or any of your favorite applications, the new ClickHouse destination for Fivetran now lets you use ClickHouse as the target database for your application data.

This is an open-source integration built over many months of hard work by our Integrations team. You can check out our release blog post here and the GitHub repository.

Other changes​

Console changes

  • Output formats support in the SQL console

Integrations changes

  • ClickPipes Kafka connector supports multi-broker setup
  • PowerBI connector supports providing ODBC driver configuration options.

April 18, 2024​

AWS Tokyo region is now available for ClickHouse Cloud​

This release introduces the new AWS Tokyo region (ap-northeast-1) for ClickHouse Cloud. Because we want ClickHouse to be the fastest database, we are continuously adding more regions for every cloud to reduce latency as much as possible. You can create your new service in Tokyo in the updated cloud console.

Other changes:

Console changes​

  • Avro format support for ClickPipes for Kafka is now Generally Available
  • Implement full support for importing resources (services and private endpoints) for the Terraform provider

Integrations changes​

  • NodeJS client major stable release: Advanced TypeScript support for query + ResultSet, URL configuration
  • Kafka Connector: Fixed a bug with ignoring exceptions when writing into DLQ, added support for Avro Enum type, published guides for using the connector on MSK and Confluent Cloud
  • Grafana: Fixed support Nullable type support in UI, fixed support for dynamic OTEL tracing table name
  • DBT: Fixed model settings for custom materialization.
  • Java client: Fixed bug with incorrect error code parsing
  • Python client: Fixed parameters binding for numeric types, fixed bugs with number list in query binding, added SQLAlchemy Point support.

April 4, 2024​

Introducing the new ClickHouse Cloud console​

This release introduces a private preview for the new cloud console.

At ClickHouse, we are constantly thinking about how to improve the developer experience. We recognize that it is not enough to provide the fastest real-time data warehouse, it also needs to be easy to use and manage.

Thousands of ClickHouse Cloud users execute billions of queries on our SQL console every month, which is why we've decided to invest more in a world-class console to make it easier than ever to interact with your ClickHouse Cloud services. Our new cloud console experience combines our standalone SQL editor with our management console in one intuitive UI.

Select customers will receive a preview of our new cloud console experience – a unified and immersive way to explore and manage your data in ClickHouse. Please reach out to us at [email protected] if you'd like priority access.

Animation showing the new ClickHouse Cloud console interface with integrated SQL editor and management features

March 28, 2024​

This release introduces support for Microsoft Azure, Horizontal Scaling via API, and Release Channels in Private Preview.

General updates​

  • Introduced support for Microsoft Azure in Private Preview. To gain access, please reach out to account management or support, or join the waitlist.
  • Introduced Release Channels – the ability to specify the timing of upgrades based on environment type. In this release, we added the "fast" release channel, which enables you to upgrade your non-production environments ahead of production (please contact support to enable).

Administration changes​

  • Added support for horizontal scaling configuration via API (private preview, please contact support to enable)
  • Improved autoscaling to scale up services encountering out of memory errors on startup
  • Added support for CMEK for AWS via the Terraform provider

Console changes​

  • Added support for Microsoft social login
  • Added parameterized query sharing capabilities in SQL console
  • Improved query editor performance significantly (from 5 secs to 1.5 sec latency in some EU regions)

Integrations changes​

Security updates​

March 14, 2024​

This release makes available in early access the new Cloud console experience, ClickPipes for bulk loading from S3 and GCS, and support for Avro format in ClickPipes for Kafka. It also upgrades the ClickHouse database version to 24.1, bringing support for new functions as well as performance and resource usage optimizations.

Console changes​

  • New Cloud console experience is available in early access (please contact support if you're interested in participating).
  • ClickPipes for bulk loading from S3 and GCS are available in early access (please contact support if you're interested in participating).
  • Support for Avro format in ClickPipes for Kafka is available in early access (please contact support if you're interested in participating).

ClickHouse version upgrade​

  • Optimizations for FINAL, vectorization improvements, faster aggregations - see 23.12 release blog for details.
  • New functions for processing punycode, string similarity, detecting outliers, as well as memory optimizations for merges and Keeper - see 24.1 release blog and presentation for details.
  • This ClickHouse cloud version is based on 24.1, you can see dozens of new features, performance improvements, and bug fixes. See core database changelogs for details.

Integrations changes​

  • Grafana: Fixed dashboard migration for v4, ad-hoc filtering logic
  • Tableau Connector: Fixed DATENAME function and rounding for "real" arguments
  • Kafka Connector: Fixed NPE in connection initialization, added ability to specify JDBC driver options
  • Golang client: Reduced the memory footprint for handling responses, fixed Date32 extreme values, fixed error reporting when compression is enabled
  • Python client: Improved timezone support in datetime parameters, improved performance for Pandas DataFrame

February 29, 2024​

This release improves SQL console application load time, adds support for SCRAM-SHA-256 authentication in ClickPipes, and extends nested structure support to Kafka Connect.

Console changes​

  • Optimized SQL console application initial load time
  • Fixed SQL console race condition resulting in 'authentication failed' error
  • Fixed behavior on the monitoring page where most recent memory allocation value was sometimes incorrect
  • Fixed behavior where SQL console sometimes issue duplicate KILL QUERY commands
  • Added support in ClickPipes for SCRAM-SHA-256 authentication method for Kafka-based data sources

Integrations changes​

  • Kafka Connector: Extended support for complex nested structures (Array, Map); added support for FixedString type; added support for ingestion into multiple databases
  • Metabase: Fixed incompatibility with ClickHouse lower than version 23.8
  • DBT: Added the ability to pass settings to model creation
  • Node.js client: Added support for long-running queries (>1hr) and handling of empty values gracefully

February 15, 2024​

This release upgrades the core database version, adds ability to set up private links via Terraform, and adds support for exactly once semantics for asynchronous inserts through Kafka Connect.

ClickHouse version upgrade​

  • S3Queue table engine for continuous, scheduled data loading from S3 is production-ready - see 23.11 release blog for details.
  • Significant performance improvements for FINAL and vectorization improvements for SIMD instructions resulting in faster queries - see 23.12 release blog for details.
  • This ClickHouse cloud version is based on 23.12, you can see dozens of new features, performance improvements, and bug fixes. See core database changelogs for details.

Console changes​

  • Added ability to set up AWS Private Link and GCP Private Service Connect through Terraform provider
  • Improved resiliency for remote file data imports
  • Added import status details flyout to all data imports
  • Added key/secret key credential support to s3 data imports

Integrations changes​

  • Kafka Connect
    • Support async_insert for exactly once (disabled by default)
  • Golang client
    • Fixed DateTime binding
    • Improved batch insert performance
  • Java client
    • Fixed request compression problem

Settings changes​

  • use_mysql_types_in_show_columns is no longer required. It will be automatically enabled when you connect through the MySQL interface.
  • async_insert_max_data_size now has the default value of 10 MiB

February 2, 2024​

This release brings availability of ClickPipes for Azure Event Hub, dramatically improves workflow for logs and traces navigation using v4 ClickHouse Grafana connector, and debuts support for Flyway and Atlas database schema management tools.

Console changes​

  • Added ClickPipes support for Azure Event Hub
  • New services are launched with default idling time of 15 mins

Integrations changes​

  • ClickHouse data source for Grafana v4 release
    • Completely rebuilt query builder to have specialized editors for Table, Logs, Time Series, and Traces
    • Completely rebuilt SQL generator to support more complicated and dynamic queries
    • Added first-class support for OpenTelemetry in Log and Trace views
    • Extended Configuration to allow to specify default tables and columns for Logs and Traces
    • Added ability to specify custom HTTP headers
    • And many more improvements - check the full changelog
  • Database schema management tools
  • Kafka Connector Sink
    • Optimized ingestion into a table with default values
    • Added support for string-based dates in DateTime64
  • Metabase
    • Added support for a connection to multiple databases

January 18, 2024​

This release brings a new region in AWS (London / eu-west-2), adds ClickPipes support for Redpanda, Upstash, and Warpstream, and improves reliability of the is_deleted core database capability.

General changes​

  • New AWS Region: London (eu-west-2)

Console changes​

  • Added ClickPipes support for Redpanda, Upstash, and Warpstream
  • Made the ClickPipes authentication mechanism configurable in the UI

Integrations changes​

  • Java client:
    • Breaking changes: Removed the ability to specify random URL handles in the call. This functionality has been removed from ClickHouse
    • Deprecations: Java CLI client and GRPC packages
    • Added support for RowBinaryWithDefaults format to reduce the batch size and workload on ClickHouse instance (request by Exabeam)
    • Made Date32 and DateTime64 range boundaries compatible with ClickHouse, compatibility with Spark Array string type, node selection mechanism
  • Kafka Connector: Added a JMX monitoring dashboard for Grafana
  • PowerBI: Made ODBC driver settings configurable in the UI
  • JavaScript client: Exposed query summary information, allow to provide a subset of specific columns for insertion, make keep_alive configurable for web client
  • Python client: Added Nothing type support for SQLAlchemy

Reliability changes​

  • User-facing backward incompatible change: Previously, two features (is_deleted and OPTIMIZE CLEANUP) under certain conditions could lead to corruption of the data in ClickHouse. To protect the integrity of the data of our users, while keeping the core of the functionality, we adjusted how this feature works. Specifically, the MergeTree setting clean_deleted_rows is now deprecated and has no effect anymore. The CLEANUP keyword is not allowed by default (to use it you will need to enable allow_experimental_replacing_merge_with_cleanup). If you decide to use CLEANUP, you need to make sure that it is always used together with FINAL, and you must guarantee that no rows with older versions will be inserted after you run OPTIMIZE FINAL CLEANUP.