**TL;DR:** ClickHouse Cloud users can now provision a fully managed Postgres service, backed by local NVMe for up to 10x faster transactions, with native CDC into ClickHouse for real-time analytics and a unified query layer via pg_clickhouse. Free until June 15, 2026, with a 50% discount after that for the duration of the beta. CDC and pg_clickhouse are included at no extra cost.

ClickHouse Cloud users can now provision a fully managed Postgres service backed by local NVMe storage, natively integrated with ClickHouse. Any ClickHouse Cloud user can now provision an enterprise-grade, fully managed Postgres service powered by local NVMe storage and natively integrated with ClickHouse. We offer a **best-of-breed data stack** that combines Postgres for transactional (OLTP) workloads and ClickHouse for analytical (OLAP) workloads, eliminating the traditional complexity of stitching together separate systems and providing a foundation essential for real-time and AI-native applications.

> [Sign up for ClickHouse Cloud today to get started with Postgres managed by ClickHouse](https://console.clickhouse.cloud/signUp?intent=PG)

You get a high-performance Postgres service backed by local NVMe with up to [10x faster]() transactional performance. Using native CDC, you can sync data from Postgres to ClickHouse in just a few clicks for [100x faster analytics](https://benchmark.clickhouse.com/). With a unified query layer powered by the [pg_clickhouse](https://clickhouse.com/blog/introducing-pg_clickhouse) extension, you can build applications that combine transactions and analytics, without managing separate systems. And all of this comes at a cost-effective price point, so you never have to compromise on a fast and reliable data foundation for building your apps.

## AI needs the “best-of-breed” data stack {#ai-needs-the-best-of-breed-data-stack}

[AI workloads are collapsing the traditional divide between transactional and analytical databases.](https://clickhouse.com/blog/ai-redrawing-database-market#real-time_analytics) Applications that once ran predictable, hard-coded queries now generate unpredictable bursts of agent-driven requests that need answers from both sides of the stack. At the same time, data volumes, concurrency, and performance expectations are growing exponentially, while security and reliability have become more critical than ever.

That’s why best-of-breed matters more than ever: Postgres for OLTP and ClickHouse for OLAP. It’s also why thousands of AI-native companies have already converged on this architecture.

![postgres_beta_may2026_image1.png](/uploads/postgres_beta_may2026_image1_4ffdd4d2c0.png)

Our vision for Postgres managed by ClickHouse is simple: make it easy  for developers to build AI-native applications on the unified data stack by eliminating the overhead of stitching together Postgres and ClickHouse through external pipelines, custom application logic, and operational complexity.

## Customers {#customers}

We announced the private preview of Postgres managed by ClickHouse earlier this year, and thousands of companies have already joined the waitlist, with many already running multi-terabyte, mission-critical production workloads.

Customers have migrated from RDS, Aurora, CloudSQL, Neon, PlanetScale Postgres, and more, while others have built entirely new AI-native applications. These workloads span cybersecurity, fintech, retail, real estate, social media, and beyond—all powered by a deeply integrated platform that unifies OLTP and OLAP with Postgres and ClickHouse.

Here are a few raw testimonials from our reference customers.

### **Physical Intelligence** {#physical-intelligence}

*Scaling AI workloads and annotation pipelines, migrated from RDS*

“ClickHouse helped us move off RDS and build a data platform that can support our growing AI workloads. We use Postgres for OLTP and ClickHouse for OLAP in the ClickHouse Cloud platform, giving researchers, training pipelines, and agents fast access to the same data foundation… As our annotation volume grows 10x and continues toward billions of annotations, ClickHouse gives us the platform headroom to keep scaling...”

### **Sterling Labs** {#sterling-labs}

*Running 8.5 TB of hot Postgres data on NVMe, migrated from Aurora.*

“Postgres managed by ClickHouse has been an incredible fit for us as we migrated from Aurora and scaled our production workload... We’re now running around 8.5 TB of hot data in Postgres, and enjoy the super low-latency offered by the NVMe drives…The performance has been simply impressive...”

### **Quinto Andar** {#quinto-andar}

*Using Postgres as a universal interface for analytics*

“With pg_clickhouse, ClickHouse becomes a plug-and-play database for virtually any third-party tool.. Instead of being forced into the overhead of BigQuery or Snowflake just to satisfy an integration like Hightouch, we can now expose key datasets directly through Postgres... It’s the best of both worlds: ClickHouse’s raw performance with the ubiquitous compatibility of Postgres.”

### **DoControl** {#docontrol}

*Simplifying cybersecurity data pipelines at scale,*

“We moved multiple multi-terabyte workloads from RDS and Aurora to Postgres managed by ClickHouse with hands-on support from the ClickHouse team.Given the scale and complexity of our cybersecurity data sources, reliability and price-performance were critical… Postgres managed by ClickHouse lets us move Postgres workloads more easily, use ClickPipes, and bring data into ClickHouse without the operational complexity we originally expected.”

Other reference customers we’d like to call out include Y Combinator companies like Trainy.ai and EndClose, AI safety companies like Mpathic, AI-native inventory management companies like Prediko, and many more power the next generation of AI-native applications on Postgres managed by ClickHouse.

## **Product** {#product}

Postgres managed by ClickHouse brings together high-performance OLTP and real-time OLAP into a single, deeply integrated platform. At the core of the platform are three foundational capabilities:

* **NVMe-backed Postgres** delivering up to 10x faster transactional performance  
* **Native CDC into ClickHouse** for real-time analytics without external pipelines  
* **pg_clickhouse**,  a unified query layer that allows applications to span transactions and analytics

Together, these capabilities eliminate the operational complexity of stitching together Postgres and analytical infrastructure manually, making it simpler to build real-time and AI-native applications.

open2_compressed.mp4

### **Fully managed migrations with ClickPipes** {#fully-managed-migrations-with-clickpipes}

Migrating production Postgres workloads is one of the hardest parts of adopting a new platform. With fully managed migration workflows powered by [ClickPipes](https://clickhouse.com/docs/cloud/managed-postgres/migrations/clickpipes), customers can move workloads from RDS, Aurora, CloudSQL, Neon, and other providers with minimal downtime and operational overhead.

Customers can continuously replicate data in real time, simplify cutovers, and avoid building custom migration infrastructure themselves, a capability that has quickly become one of the most loved aspects of the platform among customers running production workloads.

### **Enterprise-grade Postgres for production workloads** {#enterprise-grade-postgres-for-production-workloads}

Postgres managed by ClickHouse includes the operational capabilities customers expect to run mission-critical applications at scale, including:

* High availability with up to two standbys  
* Point-in-time recovery and database branching  
* Read replicas for scaling read-heavy workloads  
* 90+ PostgreSQL extensions  
* Enterprise-grade security with Private Link  
* Integrated monitoring, logs, and [Query Insights](https://clickhouse.com/blog/postgres-query-insights-clickhouse-cloud)  
* Prometheus-compatible metrics  
* Agent-based access with `clickhousectl`  
* Infrastructure as Code via OpenAPI  
* And more!

And this is just the beginning. We’re building toward a fully unified operational and analytical data platform for real-time and AI-native applications.

For detailed documentation on the above features, you can visit our official docs [here](https://clickhouse.com/docs/cloud/managed-postgres).

## Pricing {#pricing}

Postgres managed by ClickHouse is designed to be cost-effective, so developers never have to compromise on a fast, reliable data foundation powered by Postgres and ClickHouse. We’ve priced the service to deliver some of the most competitive pricing compared to alternative managed Postgres offerings. This excludes the price-performance benefits you get from local NVMe storage.

The service remains free until usage metering begins on June 15, 2026\. During Beta, all plans include a 50% discount as part of our commitment to early customers.

> For exact pricing, visit [the Pricing Calculator](https://clickhouse.com/pricing?service=postgres#pricing-calculator) to find the best configuration and pricing for your workload.

Native CDC via ClickPipes and the pg_clickhouse extension are included at no additional cost, aligning with our vision for a unified OLTP + OLAP platform powered by Postgres and ClickHouse.

The platform supports more than 50 local NVMe-backed VM configurations, ranging from 1 vCPU / 8 GB RAM / 59 GB NVMe deployments starting at approximately $32/month to clusters with 96 vCPUs / 768 GB RAM / 60 TB NVMe storage. This provides the flexibility to support everything from lightweight developer workloads to compute-intensive and storage-heavy production deployments.

During Beta, backups and network egress are also included at no additional cost. 

As we move toward General Availability, pricing and packaging may evolve. Please refer to the pricing [documentation](https://clickhouse.com/docs/cloud/managed-postgres/pricing) for additional details and disclaimers.

## Get Started {#get-started}

Postgres managed by ClickHouse is available today in Beta on ClickHouse Cloud!

Sign up for ClickHouse Cloud to provision your first NVMe-backed Postgres service, set up native CDC into ClickHouse, and start querying across both with pg_clickhouse. 

Every new account includes $300 in free credits.

Get started today

Visit the [Postgres managed by ClickHouse page](https://clickhouse.com/cloud/postgres) to learn more, or jump into the [documentation](https://clickhouse.com/docs/cloud/managed-postgres) to start building.

Postgres managed by ClickHouse is now in public beta - a fully managed, NVMe-backed Postgres service with native CDC into ClickHouse and a unified query layer via pg_clickhouse.

large_OpenHouse Blog Banner Postgres Announcement.png

medium_OpenHouse Blog Banner Postgres Announcement.png

small_OpenHouse Blog Banner Postgres Announcement.png

thumbnail_OpenHouse Blog Banner Postgres Announcement.png

OpenHouse Blog Banner Postgres Announcement.png

Postgres managed by ClickHouse is now in beta

**TL;DR:** ClickHouse Cloud ユーザーは、ローカル NVMe を活用してトランザクションを最大 10 倍高速化するフルマネージドの Postgres サービスを利用できるようになりました。リアルタイム分析向けに ClickHouse へのネイティブ CDC、さらに pg_clickhouse による統一クエリレイヤーも備えています。2026 年 6 月 15 日まで無料で、それ以降はベータ期間中 50% 割引でご利用いただけます。CDC と pg_clickhouse は追加料金なしで含まれます。

ClickHouse Cloud ユーザーは、ローカル NVMe ストレージを活用したフルマネージドの Postgres サービスを、ClickHouse とネイティブに統合された形でプロビジョニングできるようになりました。すべての ClickHouse Cloud ユーザーが、ローカル NVMe ストレージを基盤とし、ClickHouse とネイティブに統合されたエンタープライズグレードのフルマネージド Postgres サービスを利用できます。私たちは、トランザクション (OLTP) ワークロード向けの Postgres と、分析 (OLAP) ワークロード向けの ClickHouse を組み合わせた **ベスト・オブ・ブリードのデータスタック** を提供します。これにより、別々のシステムを組み合わせる従来の複雑さを排除し、リアルタイムおよび AI ネイティブなアプリケーションに不可欠な基盤を提供します。

> [今すぐ ClickHouse Cloud にサインアップして、ClickHouse 管理の Postgres を始めましょう](https://console.clickhouse.cloud/signUp?intent=PG)

ローカル NVMe を活用した高性能な Postgres サービスにより、トランザクション性能が最大 [10 倍高速]() になります。ネイティブ CDC を使えば、わずか数クリックで Postgres から ClickHouse にデータを同期し、[100 倍高速な分析](https://benchmark.clickhouse.com/) を実現できます。[pg_clickhouse](https://clickhouse.com/blog/introducing-pg_clickhouse) 拡張による統一クエリレイヤーを使えば、別々のシステムを管理することなく、トランザクションと分析を組み合わせたアプリケーションを構築できます。そしてこれらすべてが費用対効果の高い価格で提供されるため、アプリ構築のための高速かつ信頼性の高いデータ基盤について妥協する必要はありません。

## AI には「ベスト・オブ・ブリード」なデータスタックが必要 {#ai-needs-the-best-of-breed-data-stack}

[AI ワークロードは、トランザクションデータベースと分析データベースの伝統的な境界を崩しつつあります。](https://clickhouse.com/blog/ai-redrawing-database-market#real-time_analytics) かつて予測可能でハードコーディングされたクエリを実行していたアプリケーションは、今やスタックの両側から回答を必要とする、エージェント駆動の予測不可能なリクエストのバーストを生成しています。同時に、データ量、同時実行性、パフォーマンスへの期待は指数関数的に増大しており、セキュリティと信頼性はかつてないほど重要になっています。

だからこそ、ベスト・オブ・ブリードがこれまで以上に重要なのです。OLTP には Postgres、OLAP には ClickHouse。何千もの AI ネイティブ企業がすでにこのアーキテクチャに集約している理由もここにあります。

![postgres_beta_may2026_image1.png](/uploads/postgres_beta_may2026_image1_4ffdd4d2c0.png)

ClickHouse 管理の Postgres に対する私たちのビジョンはシンプルです。Postgres と ClickHouse を外部パイプライン、カスタムアプリケーションロジック、運用上の複雑さで繋ぎ合わせるオーバーヘッドを排除し、開発者が統一されたデータスタック上で AI ネイティブなアプリケーションを簡単に構築できるようにすることです。

## お客様 {#customers}

今年初めに ClickHouse 管理の Postgres のプライベートプレビューを発表したところ、すでに数千社がウェイトリストに登録し、多くがマルチテラバイト規模のミッションクリティカルな本番ワークロードを稼働させています。

お客様は RDS、Aurora、CloudSQL、Neon、PlanetScale Postgres などから移行しており、また AI ネイティブな新しいアプリケーションを一から構築している企業もあります。これらのワークロードは、サイバーセキュリティ、フィンテック、リテール、不動産、ソーシャルメディアなど多岐にわたり、いずれも Postgres と ClickHouse によって OLTP と OLAP を統合した、深く統合されたプラットフォームを基盤としています。

以下は、リファレンスカスタマーから寄せられた率直な声の一部です。

### **Physical Intelligence** {#physical-intelligence}

*AI ワークロードとアノテーションパイプラインのスケーリング、RDS から移行*

「ClickHouse は、私たちが RDS から脱却し、成長する AI ワークロードを支えるデータプラットフォームを構築するのに役立ちました。ClickHouse Cloud プラットフォーム内で OLTP には Postgres、OLAP には ClickHouse を使用しており、研究者、トレーニングパイプライン、エージェントが同じデータ基盤に高速にアクセスできるようになっています…アノテーション量が 10 倍に増加し、数十億件のアノテーションへと向かう中、ClickHouse はスケーリングを続けるためのプラットフォーム上の余裕を提供してくれます…」

### **Sterling Labs** {#sterling-labs}

*Aurora から移行し、NVMe 上で 8.5 TB のホットな Postgres データを運用*

「ClickHouse 管理の Postgres は、Aurora から移行し本番ワークロードをスケールする上で、私たちにとって素晴らしくフィットしました…現在、Postgres で約 8.5 TB のホットデータを実行しており、NVMe ドライブが提供する超低レイテンシを享受しています…パフォーマンスはまさに圧巻です…」

### **Quinto Andar** {#quinto-andar}

*分析のための汎用インターフェースとして Postgres を活用*

「pg_clickhouse があれば、ClickHouse は事実上あらゆるサードパーティツールにとってのプラグアンドプレイなデータベースになります…Hightouch のような統合のためだけに BigQuery や Snowflake のオーバーヘッドを強いられる代わりに、主要なデータセットを Postgres 経由で直接公開できるようになりました…ClickHouse の生のパフォーマンスと Postgres の遍在する互換性の両方を備えた、両方の世界の良いとこ取りです。」

### **DoControl** {#docontrol}

*大規模なサイバーセキュリティデータパイプラインの簡素化*

「ClickHouse チームの手厚いサポートを受けて、複数のマルチテラバイト規模のワークロードを RDS と Aurora から ClickHouse 管理の Postgres へ移行しました。私たちのサイバーセキュリティデータソースの規模と複雑さを考えると、信頼性とコストパフォーマンスは極めて重要でした…ClickHouse 管理の Postgres により、Postgres ワークロードをより簡単に移行し、ClickPipes を活用し、当初予想していた運用上の複雑さなしでデータを ClickHouse に取り込めるようになりました。」

その他のリファレンスカスタマーとして特筆したいのは、Trainy.ai や EndClose といった Y Combinator 企業、Mpathic のような AI セーフティ企業、Prediko のような AI ネイティブな在庫管理企業など、ClickHouse 管理の Postgres 上で次世代の AI ネイティブアプリケーションを支える数多くの企業です。

## **製品** {#product}

ClickHouse 管理の Postgres は、高性能な OLTP とリアルタイム OLAP を、深く統合された単一のプラットフォームに統合します。プラットフォームの中核には、3 つの基盤的な機能があります:

* **NVMe 基盤の Postgres** によりトランザクション性能を最大 10 倍高速化
* **ClickHouse へのネイティブ CDC** により外部パイプラインなしでリアルタイム分析を実現
* **pg_clickhouse**、アプリケーションがトランザクションと分析にまたがれる統一クエリレイヤー

これらの機能が組み合わさることで、Postgres と分析インフラを手作業で繋ぎ合わせる運用上の複雑さが解消され、リアルタイムおよび AI ネイティブなアプリケーションをよりシンプルに構築できるようになります。

### **ClickPipesによるフルマネージドな移行** {#fully-managed-migrations-with-clickpipes}

本番環境のPostgresワークロードを移行することは、新しいプラットフォームを導入するうえで最も困難な作業の1つです。[ClickPipes](https://clickhouse.com/docs/cloud/managed-postgres/migrations/clickpipes)を活用したフルマネージドの移行ワークフローにより、お客様はRDS、Aurora、CloudSQL、Neonなどの各種プロバイダーから、最小限のダウンタイムと運用負荷でワークロードを移行できます。

リアルタイムでデータを継続的にレプリケーションし、カットオーバーをシンプル化し、独自の移行インフラを構築する必要をなくすことができます。これは、本番ワークロードを運用しているお客様の間で、すでに最も支持されている機能の1つとなっています。

### **本番ワークロード向けのエンタープライズグレードPostgres** {#enterprise-grade-postgres-for-production-workloads}

ClickHouseが管理するPostgresには、ミッションクリティカルなアプリケーションを大規模に運用するためにお客様が期待する運用機能が含まれています。たとえば次のようなものです。

* 最大2つのスタンバイによる高可用性
* ポイントインタイムリカバリとデータベースブランチング
* 読み取り負荷の高いワークロードをスケールさせるリードレプリカ
* 90以上のPostgreSQL拡張機能
* Private Linkによるエンタープライズグレードのセキュリティ
* 統合されたモニタリング、ログ、および[Query Insights](https://clickhouse.com/blog/postgres-query-insights-clickhouse-cloud)
* Prometheus互換のメトリクス
* `clickhousectl`によるエージェントベースのアクセス
* OpenAPIによるInfrastructure as Code
* その他多数!

そして、これはまだ始まりに過ぎません。私たちは、リアルタイムおよびAIネイティブのアプリケーション向けに、運用系と分析系を完全に統合したデータプラットフォームの構築を目指しています。

上記の機能に関する詳細なドキュメントは、[こちら](https://clickhouse.com/docs/cloud/managed-postgres)の公式ドキュメントをご参照ください。

## 料金 {#pricing}

ClickHouseが管理するPostgresは、コスト効率を重視して設計されており、開発者はPostgresとClickHouseによる高速で信頼性の高いデータ基盤を妥協なく利用できます。本サービスは、他のマネージドPostgresサービスと比較しても非常に競争力のある価格設定となっています。これには、ローカルNVMeストレージによる価格性能上のメリットは含まれていません。

本サービスは、2026年6月15日に利用量メータリングが開始されるまで無料です。Beta期間中は、早期にご利用いただくお客様への感謝の意を込めて、すべてのプランで50%の割引が適用されます。

> 正確な料金については、[料金計算ツール](https://clickhouse.com/pricing?service=postgres#pricing-calculator)にアクセスして、ワークロードに最適な構成と料金をご確認ください。

ClickPipesによるネイティブCDCおよびpg_clickhouse拡張機能は追加料金なしで提供され、PostgresとClickHouseによる統合OLTP + OLAPプラットフォームというビジョンに沿ったものとなっています。

本プラットフォームは、1 vCPU / 8 GB RAM / 59 GB NVMe(月額約32ドルから)の構成から、96 vCPU / 768 GB RAM / 60 TB NVMeストレージのクラスタまで、50を超えるローカルNVMeバックアップ型のVM構成をサポートします。これにより、軽量な開発者向けワークロードから、コンピューティング集約型およびストレージ集約型の本番デプロイメントまで、柔軟に対応できます。

Beta期間中は、バックアップとネットワーク送信(egress)も追加料金なしで提供されます。

一般提供(General Availability)に向けて、料金やパッケージは変更される可能性があります。詳細および免責事項については、料金[ドキュメント](https://clickhouse.com/docs/cloud/managed-postgres/pricing)をご参照ください。

## はじめる {#get-started}

ClickHouseが管理するPostgresは、本日よりClickHouse CloudでBeta提供を開始しました!

ClickHouse Cloud にサインアップして、最初の NVMe ベース Postgres サービスをプロビジョニングし、ClickHouse へのネイティブ CDC をセットアップして、pg_clickhouse で両者をまたいだクエリを開始しましょう。

新規アカウントにはすべて $300 分の無料クレジットが含まれます。

今すぐ始める

詳細については [ClickHouse によるマネージド Postgres ページ](https://clickhouse.com/cloud/postgres) をご覧いただくか、[ドキュメント](https://clickhouse.com/docs/cloud/managed-postgres) を参照して構築を始めてください。

ClickHouseが管理するPostgresがパブリックベータとして利用可能になりました。NVMeバックエンドのフルマネージドPostgresサービスで、ClickHouseへのネイティブCDCとpg_clickhouseによる統合クエリレイヤーを備えています。

ClickHouseが管理するPostgresがベータ版で提供開始

As we build [Postgres managed by ClickHouse](https://clickhouse.com/cloud/postgres), an NVMe-backed Postgres service native integrated with ClickHouse, our top priority (P0) is delivering an always-available, reliable, and operationally observable managed Postgres offering. We believe this is table stakes for OLTP workloads!

With that vision in mind, we’ve been investing heavily in state-of-the-art features to ensure customers get an enterprise-grade Postgres experience. In this blog, we highlight several key features we’ve shipped over the past few months and share a preview of what’s ahead on our roadmap to further strengthen platform maturity and enterprise readiness. Let’s get started!

## Cross AZ HA

Postgres managed by ClickHouse supports up to two synchronous standbys spread across availability zones, protecting you from node, rack, or full AZ failures. With 2 standbys configured, we use [quorum-based streaming replication](https://www.postgresql.org/docs/current/runtime-config-replication.html#GUC-SYNCHRONOUS-STANDBY-NAMES): a write is acknowledged when at least one standby confirms it. This provides strong durability guarantees, without turning every commit into a latency tax. In the event of a failure, failover is automatic, promoting a standby with minimal disruption.

![quorum-based streaming replication](/uploads/552707022_938b77d4_1c8d_4fae_ab97_c0ed6edacee2_92b4642245.jpg)

For cost-sensitive workloads that require high availability, but aren’t tier 0, you have the flexibility to choose 1 cross-AZ standby. It uses synchronous replication to provide strong availability, while offering lesser fault tolerance than a two-standby quorum configuration.

> An important architectural nuance: our HA standbys are not exposed for reads, ensuring that we prioritize failover readiness and data durability over opportunistic read scaling.

HA standbys are reserved strictly for failover and are not exposed for read traffic. Read traffic on standbys can compete with WAL replay, increasing replication lag, and can delay failover readiness. Long-running queries on replicas can also interfere with VACUUM and bloat control on primary. We avoid those trade-offs by keeping standbys focused solely for HA. If you need read scaling, **we provide separate read replicas designed specifically for that purpose**.

## Reliable CDC to ClickHouse (failover-safe slots)

The service comes with native ClickHouse integration with built-in Change Data Capture (CDC), enabling continuous replication of transactional data into ClickHouse for real-time analytics. The integration is powered by ClickPipes/PeerDB, a battle tested replication engine supporting 100s of Postgres customers.

> A unique reliability feature of the service is built-in failover replication slots, preventing resyncs on primary failover.

In most managed Postgres services, logical replication slots are tied to the primary instance. During high-availability failovers, maintenance events, or scaling operations, these slots can be lost or require manual recreation, interrupting CDC pipelines and potentially forcing full re-syncs.

Postgres by ClickHouse includes built-in infrastructure for failover replication slots when syncing data to ClickHouse. These slots are preserved across HA failovers and scaling operations. As a result, CDC pipelines continue running without manual intervention or slot re-creation when the primary changes, reducing the risk of costly resyncs on large databases.

## Backups, PITR and Forks

High availability protects you from infrastructure failures. Backups protect you from everything else.

Every Postgres service includes **automatic backups** with support for **point-in-time recovery (PITR)** and **forks**. We use **WAL-G**, a widely adopted open-source tool, to take full base backups and continuously archive WAL to object storage (S3-compatible). WAL-G is well tuned to perform full backups and restores, as well as WAL archival and retrieval, in parallel to meet the high throughput demands of large-scale workloads. We also use the [`wal-g` daemon](https://github.com/wal-g/wal-g/blob/master/docs/PostgreSQL.md#daemon), as it runs as a persistent process, eliminating per-WAL process startup overhead and enabling efficient, low-latency, and reliable WAL shipping under sustained write volumes.

backups_e.mp4


Continuous WAL archiving allows you to restore your database to an exact timestamp within the retention window (default 7 days), not just to the last snapshot. Forks are built on the same mechanism. You can create a new Postgres service from any recovery point, making it easy to test migrations, debug issues, or run backfills without impacting production.

Together with HA, backups provide a second layer of protection, covering not just infrastructure failures, but human error and application bugs as well.

## Zero compromise security

Postgres managed by ClickHouse is designed with practical, production-grade security controls across network access, encryption, and isolation.

> A key architectural differentiator is that **every Postgres cluster runs inside its own dedicated, isolated VPC.** 

This provides hard network-level isolation between customer environments, with no shared runtime and no cross-database interference. The result is strong blast-radius containment and reduced multi-tenant risk. While many managed database services rely on container orchestration platforms with shared networking layers, we provision dedicated environments per cluster to deliver a higher level of separation by default.

![dedicated, isolated VPCs](/uploads/552707023_277c2c51_35a9_4343_9a77_77ee0d1edca1_14eec91d6a.jpg)

At the network layer, access can be restricted using [IP allow-listing](https://clickhouse.com/docs/cloud/managed-postgres/security#ip-whitelisting), ensuring that only approved application servers, VPN endpoints, bastion hosts, or CI/CD systems can connect to the database.

All data is [encrypted](https://clickhouse.com/docs/cloud/managed-postgres/security#encryption) both in transit and at rest. Client connections are secured with TLS (SSL) to protect against interception and man-in-the-middle attacks. Encrypted connections can be strictly enforced, and for environments requiring stronger authentication guarantees, certificate-based client verification can be used. Data at rest, including NVMe storage and object storage used for backups and WAL archives,  is encrypted using AES-256, the industry-standard encryption algorithm widely adopted across cloud providers. Encryption keys are securely managed by the underlying cloud infrastructure.

For environments requiring stronger network isolation, [Private Link support](https://clickhouse.com/docs/cloud/managed-postgres/security#private-link) enables private connectivity between your VPC and the Postgres service, ensuring that database traffic does not traverse the public internet.

## Monitoring

The Postgres service comes with built-in, out-of-the-box metrics that provide visibility into both system health and database performance. Today, it includes 10+ core metrics, allowing you to monitor CPU usage, memory consumption, disk I/O, storage growth, and network traffic, alongside Postgres-specific signals such as connections, transaction throughput, cache hit ratios, row activity, and deadlocks.

monitoring_e.mp4

We’re actively expanding operational visibility, including Query Performance Insights and more comprehensive logging.

## Looking Ahead

As we execute on **our vision of delivering a unified data stack for OLTP and OLAP by combining Postgres and ClickHouse**, providing an enterprise-grade experience is a given!

In the coming quarters, we’re expanding the platform’s enterprise readiness across reliability, performance, and operational depth. A few features include support for additional cloud providers (GCP), CDC v2 for close to sub-second replication into ClickHouse, and Query Performance Insights (QPI) for deeper visibility into query behavior and database internals.

We’re also investing in core platform capabilities such as storage autoscaling, OpenAPI and Terraform support for IaC, one-click fully managed Postgres-to-Postgres migrations, improved operational visibility (e.g., scaling progress), a built-in intuitive SQL console and more.

Stay tuned.

ClickHouse + Postgres has become the unified data stack for applications that scale. With Managed Postgres now available in ClickHouse Cloud, this stack is a day-1 decision.

Try Postgres managed by ClickHouse

We’ve been investing heavily in state-of-the-art features to ensure customers get an enterprise-grade Postgres experience.

large_552217859-2606de73-2445-4ff5-886c-27321950cc73.jpg

medium_552217859-2606de73-2445-4ff5-886c-27321950cc73.jpg

small_552217859-2606de73-2445-4ff5-886c-27321950cc73.jpg

thumbnail_552217859-2606de73-2445-4ff5-886c-27321950cc73.jpg

552217859-2606de73-2445-4ff5-886c-27321950cc73.jpg

Building an enterprise Postgres service in ClickHouse Cloud

Today, we're announcing a fast, scalable and an enterprise-grade managed Postgres service natively integrated with ClickHouse, [built for real-time and AI-driven applications](https://clickhouse.com/resources/engineering/managed-postgres-for-ai-and-real-time-apps). We are offering a **Unified Data Stack** that brings transactional and analytical workloads together without traditional complexity. By combining Postgres for transactions with ClickHouse for analytics, teams get best-in-class performance and scalability on an open-source foundation while significantly reducing operational overhead.

Developers get a high-performance, scalable Postgres service backed by local NVMe storage, delivering up to 10X faster performance for fast-growing workloads that are mostly disk-bound. In just a few clicks, using native CDC capabilities, they can sync transactional data in Postgres to ClickHouse to make it analytics-ready, unlocking up to [100X faster analytics](https://benchmark.clickhouse.com/#system=+lik%7CgQ&type=-&machine=-ca2l%7C6t%7Cg4e%7C6ax%7Cae-l%7C6ale%7Cg-l%7C3al&cluster_size=-&opensource=-&hardware=-&tuned=+n&metric=combined&queries=-). With a unified query layer powered by the [pg_clickhouse](https://github.com/ClickHouse/pg_clickhouse) Postgres extension, they can build applications that seamlessly span transactions and analytics without managing separate systems.

To try ClickHouse's native Postgres service, sign up for Private Preview using this link.

Get started with our native Postgres Service

This demo showcases how the various pieces come together to form a Unified Data Stack powered by Postgres and ClickHouse.

<iframe width="768" height="432" src="https://www.youtube.com/embed/rpBA13nQxAk?si=Xu1e9FERMcnXgKWh" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

To achieve this, we're partnering with [Ubicloud](https://www.ubicloud.com/), an open-source cloud company that offers one of the fastest Postgres services and whose [founding team](https://www.ubicloud.com/about-ubicloud) has a track record of delivering world-class Postgres at Citus Data, Heroku, and Microsoft.


> Noah Pryor, the CTO of [Beehiiv](https://www.beehiiv.com/), a large ClickHouse customer, shared excitement for this launch, "We’re excited to see ClickHouse entering the Postgres ecosystem. At Beehiiv, we rely heavily on both Postgres and ClickHouse to power our mission-critical, customer-facing applications. We’ve invested significant effort in integrating these technologies, so a tighter, more native integration between them would materially simplify our architecture."

## Why Postgres managed by ClickHouse

### Postgres, built for performance and scale

Across thousands of fast growing AI-driven companies using Postgres, performance and scalability consistently emerge as challenges. Common issues include slower ingestion (mainly updates), slower vacuums, increased tail latency for reads and writes, long-running, slow transactions causing lock contention and sudden WAL spikes, among others. In most cases, these issues stem from limited disk IOPS and suboptimal disk latency. With unlimited IOPS, Postgres can operate more efficiently than it does today on most other managed Postgres services.

Postgres managed by ClickHouse is an enterprise-grade managed Postgres service backed by NVMe storage. The NVMe storage is physically colocated with compute, delivering orders of magnitude lower disk latency and higher IOPS than alternatives such as EBS-backed volumes, which require a network round trip for disk access. As a result, disk access latency can be on the order of microseconds rather than milliseconds. For fast-growing Postgres workloads that are primarily throttled by disk IOPS and latency, we expect up to 10X speedups at the same cost and hardware profile compared to other Postgres providers.

> *"We're excited that ClickHouse is offering a native Postgres service to complement its core OLAP database. NVMe-backed Postgres performance, combined with seamless integration with CDC and pg_clickhouse for fast analytics, makes it very compelling for us to build our fast-growing applications."* - Aditya Maru, Co-founder, [Blacksmith.sh](http://blacksmith.sh)

The image below shows the reduction in P99 latency for disk-bound workloads when moving from traditional Postgres to NVMe-backed Postgres.

![postgres-launch-1.png](/uploads/postgres_launch_1_5e02d22e1a.png)

#### Ubicloud's NVMe-backed Postgres stood out

To select the best architecture, we evaluated many Postgres providers as potential partners. Ubicloud stood out by offering a reliable Postgres service backed by NVMe storage that addressed the same customer challenges we observed and outperformed other providers by orders of magnitude in real-world benchmarks. We plan to share a detailed blog post on our evaluation process in the future.

We evaluated alternative architectures, including separating storage and compute, and found that object-storage-backed Postgres is not well suited for OLTP workloads, where consistent  and predictable latency and throughput are essential. Object storage works well for analytics, but its higher latency and coarse-grained I/O make it challenging to support transactional patterns. We’ll dive deeper into this trade-off in a follow-up blog post.

### Deeply integrated with ClickHouse

Developers can natively integrate their Postgres with ClickHouse to enable 100x faster analytics. It comes with two critical features enabling this integration. These features are central to powering the Unified Data Stack, combining Postgres for transactions and ClickHouse for analytics with no overhead.

The diagram below depicts the Unified Data Stack, highlighting Postgres for OLTP, ClickHouse for OLAP, continuous replication between the two, and a unified query layer via the `pg_clickhouse` extension. 

![postgres-launch-2.png](/uploads/postgres_launch_2_5b8721a622.png)

### Postgres to ClickHouse replication

Seamlessly replicate Postgres data to ClickHouse in real time to make it analytics-ready. We support both an initial load to migrate existing data to ClickHouse and CDC for incremental syncs. This feature is powered by the Postgres CDC connector in [ClickPipes](https://clickhouse.com/docs/integrations/clickpipes/postgres), which is battle-tested by 100s of enterprise customers moving 100s of TB per month.

We plan to offer additional features for more native CDC, including sub-second replication latency, replication of ongoing transactions using logical replication v2 for reliable slot flushing, etc. all of which will be exclusive to our native Postgres service.

### pg_clickhouse: unified query layer for OLTP and OLAP

Every Postgres service comes with the [pg_clickhouse](https://clickhouse.com/blog/introducing-pg_clickhouse) Postgres extension, which enables users to query ClickHouse directly from Postgres. It allows Postgres to act as a unified query layer, powering not just transactions but also analytics using ClickHouse internally, making it easy for developers to build applications.

`pg_clickhouse` provides comprehensive query pushdown to ClickHouse for efficient query execution, including support for FILTERs, JOINs, SEMI-JOINs, aggregations, functions and more. Currently, 14 of 22 TPC-H queries are fully pushed down, delivering over 60x performance improvements compared to standard Postgres.

We will continue expanding pushdown support to more complex queries, including CTEs, window functions, and beyond, all with the goal of enabling users to build fast analytics from the Postgres layer itself while still leveraging the power of ClickHouse.

Users can seamlessly create foreign tables in Postgres that point to ClickHouse tables under the hood and start powering analytics using these foreign tables. As we evolve `pg_clickhouse`, we'll work toward making this experience as UX-friendly as possible, including auto-creating foreign tables for replicated data, auto-routing transactions and analytics to the appropriate engines (Postgres or ClickHouse), and more.

### Enterprise-grade Postgres

#### Reliability without compromise

Our native Postgres service provides enterprise-grade reliability from day one. We offer strong high availability (HA), allowing a primary with up to two standby instances across different availability zones (AZs) using quorum-based replication. These standby instances are dedicated to HA and auto-failover; they are not exposed as read replicas, as that could impact HA guarantees. Read scaling is handled via separate read replicas (details [here](https://clickhouse.com/docs/cloud/managed-postgres/read-replicas)).

For disaster recovery, every service comes with automatic backups that support forks and point-in-time recovery. The backups run on [WAL-G](https://github.com/wal-g/wal-g), a well-known open-source Postgres tool that handles full backups and continuous WAL archiving to object storage like S3. WAL-G was [originally created](https://www.citusdata.com/blog/2017/08/18/introducing-wal-g-faster-restores-for-postgres/) by Daniel Farina, co-founder of Ubicloud, during his time at Citus Data, another testament to the quality of the team we're partnering with.

#### Built on ClickHouse Cloud trust

It is built to meet the same high standards for security, privacy, and compliance as ClickHouse Cloud. As part of a unique partnership, Ubicloud runs entirely within AWS accounts owned by ClickHouse Cloud and is subject to the same security, privacy, and governance controls applied across the platform. It already includes core capabilities such as SAML/SSO, IP allow-listing, data encryption at rest and in transit, as well as PrivateLink (available via a support ticket), with plans to further align and unify with the advanced features offered by ClickHouse Cloud over time.

### Open source first

The managed Postgres service is built on an open source first philosophy. [Postgres](https://github.com/postgres/postgres) and [ClickHouse](https://github.com/ClickHouse/ClickHouse) are both open-source databases with large, thriving communities. Every component that integrates Postgres and ClickHouse is open source, from the [pg_clickhouse](https://github.com/ClickHouse/pg_clickhouse) extension that enables a unified query layer to [PeerDB](https://github.com/PeerDB-io/peerdb), which powers real-time CDC and replication.

[Ubicloud](https://github.com/ubicloud/ubicloud), our strategic partner on the managed Postgres service, is also an open source company, deeply aligned with this philosophy This was a major factor in choosing the right partner. We are jointly contributing to the open-source ecosystem to deliver a world-class Postgres to the community, and this collaboration is already underway. [Here](https://github.com/ubicloud/ubicloud/pulls?q=is%3Apr+author%3Aserprex) are a few key contributions from the ClickHouse team to the Ubicloud open-source repository.

This open source foundation ensures there is no vendor lock-in, giving developers full control, transparency, and long-term flexibility over their data stack.

## Try Postgres managed by ClickHouse



Our team will reach out within a day and work closely with you to provide early access. **The private preview is fully free and entails no cost.**

It is currently available on AWS in 10 public regions, with over 50 NVMe-backed configurations ranging from 2 vCPUs / 8 GB RAM / 118 GB disk to 96 vCPUs / 768 GB RAM / 60 TB storage. We plan to expand to additional regions and other cloud service providers, including GCP and Azure, in the near future. The offering includes all standard managed service features, including high availability, forks, read replicas, PgBouncer, upgrades, and more, along with deep integration with ClickHouse.

ClickHouse launch a fast, scalable and enterprise-grade managed Postgres service natively integrated with ClickHouse, built for real-time and AI-driven applications.

large_postgres-launch-banner.png

medium_postgres-launch-banner.png

small_postgres-launch-banner.png

thumbnail_postgres-launch-banner.png

postgres-launch-banner.png

Fast, scalable, enterprise-grade Postgres natively integrated with ClickHouse

Seamlessly connect your external data sources to ClickHouse Cloud - try out ClickPipes today!

A look back at the evolution of the Postgres CDC connector in ClickHouse Cloud over its first year, following the 2024 PeerDB acquisition. 

large_Postgres CDC ClickHouse Review.jpg

medium_Postgres CDC ClickHouse Review.jpg

small_Postgres CDC ClickHouse Review.jpg

thumbnail_Postgres CDC ClickHouse Review.jpg

Postgres CDC ClickHouse Review.jpg

Postgres CDC in ClickHouse, A year in review

本日、ClickPipes向け [Postgres CDCコネクタ](https://clickhouse.com/cloud/clickpipes/postgres-cdc-connector)の一般提供開始を発表できることを嬉しく思います。これにより、お客様はPostgresデータベースをClickHouse Cloudへ数クリックで簡単にレプリケートできるようになります。

この発表は、大きなマイルストーンです。[PeerDB](https://www.peerdb.io/) ― Postgres CDCに特化した企業であり、昨年[ClickHouseが統合](https://clickhouse.com/blog/clickhouse-welcomes-peerdb-adding-the-fastest-postgres-cdc-to-the-fastest-olap-database)したパートナー ― は、ClickPipes内のPostgres CDCコネクタとしてClickHouse Cloudに完全統合されました。このコネクタは、エンタープライズレベルのPostgresユースケースにも対応可能です。

> [ClickHouse Cloudにサインアップ](https://console.clickhouse.cloud/signup)して、[ClickPipes向けPostgres CDCコネクタ](https://clickhouse.com/docs/integrations/clickpipes/postgres)をお試しください。 

## Postgres + ClickHouse = 「デフォルトのデータスタック」

近年、[GitLab](https://about.gitlab.com/blog/2022/04/29/two-sizes-fit-most-postgresql-and-clickhouse/)、[CloudFlare](https://blog.cloudflare.com/http-analytics-for-6m-requests-per-second-using-clickhouse...)、[Instacart](https://tech.instacart.com/real-time-fraud-detection-with-yoda-and-clickhouse-bd08e9dbe3f4) などの企業をはじめ、多くのビジネスで共通のパターンが見られるようになっています。それは、PostgresとClickHouseを組み合わせて、あらゆるデータ課題を解決するというものです。

* **Postgres** はトランザクション型Webアプリケーションのためのシステム・オブ・レコードとして機能します。
* **ClickHouse** はリアルタイム分析とレポート用途のためのシステム・オブ・アナリシスとして機能します。

このパターンはAI時代に入りさらに加速しており、[LangChain](https://clickhouse.com/blog/langchain-why-we-choose-clickhouse-to-power-langchain)、[LangFuse](https://langfuse.com/blog/2024-12-langfuse-v3-infrastructure-evolution)、[Vapi](https://neon.tech/blog/vapi-voice-agents-neon) などの企業も同様のアーキテクチャを採用しています。私たちは、Postgres + ClickHouseがモダンビジネスにおける「標準的なデータスタック」になりつつあると考えています。その統合を“魔法のようにシームレス”にすることが私たちの目標です。そして今回のPostgres CDCコネクタは、そのための**最初の大きなステップ**です。PostgresのデータをClickHouseへ簡単に取り込むことで、リアルタイム分析をスムーズに実現できるようになります。

以下は、PostgresとClickHouseをPostgres CDCで組み合わせた際のリファレンスアーキテクチャの図です。Postgresは低レイテンシなトランザクションを処理し、ClickHouseが高速な分析を支えます。

<img preview="/uploads/Postgres_Click_House_reference_architecture_61fe52239a.png"  src="/uploads/Postgres_Click_House_reference_architecture_61fe52239a.png" alt="catalogue_lakehouse.png" class="h-auto w-auto max-w-full"  style="width: 100%;">

## メトリクスと導入企業の事例

過去6か月間、このコネクタはユーザーフィードバックをもとに急速に進化を遂げながら、広範なベータフェーズを経てきました。現在では、**数百のミッションクリティカルなワークロード**を支え、**月間100TB以上のデータ**をClickHouseに移行しています。

代表的な導入企業には、米国の大手採用プラットフォーム [**Ashby**](https://www.ashbyhq.com/)、急成長中のサイバーセキュリティスタートアップ [**Seemplicity**](https://seemplicity.io/)、米国有数の自動車小売企業 [**AutoNation**](https://www.autonation.com/) などがあります。以下は、一部のお客様からの声です。

> 「ClickHouseはAshbyの顧客向け分析基盤として稼働しており、完全に動的なインサイトを超高速で提供しています。一方でPostgresはコアトランザクションを処理しています。ClickPipes経由のPostgres CDCにより、テラバイト規模のデータをシームレスにレプリケートでき、リアルタイム分析がさらに高速化しました。以前は数分かかっていたレポートが、今では1秒以内で完了します。エンタープライズ顧客がリアルタイムで意思決定する際に活用するダッシュボードは、まさに“真実の情報源”です。PostgresとClickHouseを組み合わせることで、私たちは信頼性の高い体験と、スケールに応じたデータ処理を提供できています。」
> — [Elenie Godzaridis](https://www.linkedin.com/in/elenie-godzaridis/), Director of Engineering, [Ashby](https://www.ashbyhq.com/)

> 「最初はDebeziumを使って社内でPostgres CDCを実装しようとしましたが、あまりに複雑すぎました。PostgresのビットをClickHouseに変換することを使命にしているエンジニアたちが作ったマネージド製品のほうが、自分たちの手で作るものよりも優れていると確信していました。」
> — [Tal Shargal](https://www.linkedin.com/in/tal-shargal-29388671/), Chief Architect, [Seemplicity](https://seemplicity.io/), [*導入事例はこちら*](https://clickhouse.com/blog/seemplicity-scaled-real-time-security-analytics-with-postgres-cdc-and-clickhouse)

![Chart.png](https://cms.clickhouse-dev.com:1337/uploads/Chart_7c2a541127.png)

**このグラフは、過去1年間におけるPostgres CDCからClickHouseへの使用量の成長を示しています。** [**リファレンス**](https://www.linkedin.com/feed/update/urn:li:activity:7310749593296588800/)

## 製品の特長

<iframe width="660" height="371" src="https://www.youtube.com/embed/s1hREjN02S0?si=J2ByKIkTTU9Mw94m" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

Postgres CDCコネクタは、PostgresとClickHouseのために専用設計されたコネクタであり、豊富な機能を備えています。以下は特に注目すべき機能の一部です：

1. **初回ロードおよび再同期が10倍高速化** –  [並列スナップショット](https://blog.peerdb.io/parallelized-initial-load-for-cdc-based-streaming-from-postgres)により、大規模な単一テーブルを並列に読み込み可能。テラバイト規模のデータでも、従来の「数日」ではなく「数時間」で移行できます。

2. **最小10秒のレプリケーション遅延** –  レプリケーションスロットは[再接続なしで継続的に消費](https://clickhouse.com/blog/enhancing-postgres-to-clickhouse-replication-using-peerdb#efficiently-flush-the-replication-slot)され、ClickHouseの[ReplacingMergeTree](https://clickhouse.com/blog/postgres-to-clickhouse-data-modeling-tips-v2)がエンドツーエンドの遅延を最小化します。

3. **スキーマ変更の自動レプリケーション** –  [ADD COLUMNやDROP COLUMNといった操作](https://clickhouse.com/docs/integrations/clickpipes/postgres/schema-changes)にも対応し、手動による対応は不要です。

4. **テーブル・カラム単位の除外設定** –  PII（個人情報）保護やネットワークスループットの最適化のために、きめ細かな制御が可能です。

5. **セキュアな接続** –  [**AWS PrivateLink**](https://clickhouse.com/docs/integrations/clickpipes/aws-privatelink) や [**SSHトンネリング**](https://clickhouse.com/docs/integrations/clickpipes/postgres#optional-setting-up-ssh-tunneling) に対応しており、ソースのPostgresデータベースと安全かつプライベートに接続できます。

6. **Postgresのネイティブ機能に対応** –  [パーティションテーブル](https://blog.peerdb.io/real-time-change-data-capture-for-postgres-partitioned-tables)、TOASTカラム、[配列やJSONなどの高度なデータ型](https://clickhouse.com/docs/integrations/clickpipes/postgres/faq#how-are-postgres-data-types-mapped-to-clickhouse)のレプリケーションもサポートしています。

7. **Open API（ベータ）** –  [APIやCLIによるパイプの作成・管理](https://clickhouse.com/docs/integrations/clickpipes/postgres/faq#can-clickpipe-creation-be-automated-or-done-via-api-or-cli)が可能で、ClickHouse Cloud + ClickPipes環境を個別に構築・運用するISV（独立系ソフトウェアベンダー）にも最適です。[**Terraform**](https://registry.terraform.io/providers/ClickHouse/clickhouse/3.2.0-alpha1/docs/resources/clickpipe) **によるIaC対応も近日提供予定です。**

## 料金について

一般提供（GA）の開始に伴い、[ベータ期間中にもお知らせしていた](https://clickhouse.com/blog/postgres-cdc-connector-clickpipes-public-beta#pricing)とおり、Postgres CDCコネクタの料金体系を導入します。私たちの目標は、PostgresデータベースをClickHouseにシームレスかつ手頃な価格で接続できるようにするというビジョンを保ちながら、非常に競争力のある価格を実現することです。

このコネクタは、外部のETLツールや他のデータベースプラットフォームにある類似機能と比べて、**5倍以上のコスト効率**を誇ります。

なお、Postgres CDC ClickPipesを利用している**すべてのお客様（既存・新規）に対して、2025年9月1日より月次課金が開始**されます。必要に応じてコストを最適化するための**3か月間の猶予期間**が設けられていますが、多くのお客様にとって設定の変更は不要と想定しています。料金体系の詳細、具体的な例、よくある質問については[こちら](https://clickhouse.com/docs/cloud/manage/billing/overview#clickpipes-for-postgres-cdc)をご覧ください。

## Postgres CDC を使って ClickHouse にデータを取り込む方法

以下のリンクから、PostgresデータベースをClickHouse Cloudに接続して、超高速な分析を体験してみてください。

* [PostgresからClickHouseへのデータ取り込み（CDCを使用）](https://clickhouse.com/docs/en/integrations/clickpipes/postgres)

* [ClickPipes for Postgres よくある質問（FAQ）](https://clickhouse.com/docs/en/integrations/clickpipes/postgres/faq)

* [ClickHouse Cloudを無料で試す](https://clickhouse.com/docs/en/cloud/get-started/cloud-quick-start)



ClickPipes向けPostgres CDCコネクタの一般提供を開始しました。ClickHouse Cloudにネイティブに統合されており、Postgresデータベースのレプリケーションを数クリックで高速に実現できます。

Sai Srirampur

Blog posts by Sai Srirampur

Postgres managed by ClickHouse is now in beta

ClickHouseが管理するPostgresがベータ版で提供開始

Building an enterprise Postgres service in ClickHouse Cloud

Fast, scalable, enterprise-grade Postgres natively integrated with ClickHouse

Postgres CDC in ClickHouse, A year in review

ClickPipes向けPostgres CDCコネクタが一般提供開始