ClickHouse and Datadog Move Observability Toward a Storage-Agnostic Future
ClickHouse and Datadog announced a new product partnership on June 9, 2026 that allows organizations to route logs from Datadog Observability Pipelines directly into ClickHouse and search those logs from Datadog Log Explorer without re-ingesting the data. The integration launches in preview and introduces 2 core capabilities: a native ClickHouse destination inside Datadog Observability Pipelines and federated search for ClickHouse-hosted logs through Datadog Log Explorer.
ClickHouse is a high-performance analytical database headquartered in the San Francisco Bay Area and increasingly used for large-scale observability workloads. Datadog operates one of the world's largest observability platforms, serving enterprise software and cloud infrastructure teams globally. Together, the companies are attempting to address a growing problem facing engineering teams: how to retain more telemetry data without sacrificing usability or driving costs higher.
The announcement reflects a broader shift across infrastructure software. Observability is becoming more modular. Storage, search, routing, and analysis are increasingly being separated into distinct layers rather than bundled inside a single platform.
What Happened
Engineering teams have spent years trapped between 2 uncomfortable options: store everything and watch costs climb, or store less and discover the missing log was the one that explained the outage. That tension has only intensified as AI agents, LLM applications, distributed services, and machine-generated workflows generate larger volumes of telemetry data. Infrastructure teams are collecting more information than ever, but retaining and searching that information remains an expensive balancing act.
The new ClickHouse and Datadog partnership is designed to change that equation. Datadog customers can now use a native ClickHouse destination within Observability Pipelines to route logs directly into ClickHouse. During transit, Datadog can still perform parsing, enrichment, filtering, and redaction before logs arrive at their final destination.
The second capability may prove even more important. Through federated search, engineers can search logs stored in ClickHouse directly from Datadog Log Explorer while the data remains in ClickHouse. No duplication. No secondary ingestion. No requirement to move data back into Datadog simply to investigate an incident. That distinction matters because storage and investigation have historically been tightly linked inside observability platforms, and this integration begins separating those functions.
For readers seeking primary documentation, both companies published official announcements detailing the launch, including ClickHouse's official partnership announcement and Datadog's corresponding product release.
About ClickHouse
ClickHouse occupies an increasingly important position inside modern Developer Infrastructure and Data Infrastructure. The company traces the origins of its database technology back to 2009, with the project becoming open source in 2016. ClickHouse was commercialized through ClickHouse, Inc., which was incorporated in 2021 and is headquartered in the San Francisco Bay Area.
The company is led by Co-Founder and CEO Aaron Katz, alongside Co-Founder and CTO Alexey Milovidov and Co-Founder and President Yury Izrailevsky. Originally known for analytical workloads, ClickHouse has expanded aggressively into observability and AI-related infrastructure. Organizations including OpenAI, Anthropic, DoorDash, and Shopify have publicly been identified by ClickHouse as users of the platform for large-scale observability analytics.
Recent momentum has been substantial. Bloomberg reported that ClickHouse reached a $15B valuation following a $400M financing round in January 2026. TechCrunch later reported that the company surpassed $250M in annualized revenue while continuing to expand its enterprise footprint.
The acquisition of Langfuse in 2026 also pushed ClickHouse deeper into the rapidly growing AI Observability market.
Why This Matters
Most infrastructure announcements promise speed. This one is really about economics. Telemetry growth has become relentless. Every application generates logs. Every service emits metrics. Every AI workflow creates additional traces, events, and operational records. The problem is not collecting data. The problem is deciding where that data should live.
Traditional observability models often force customers into a single storage strategy. The more data retained, the larger the bill becomes. Teams eventually start sampling logs, shortening retention periods, or making compromises they hope never become relevant during an incident.
ClickHouse and Datadog are proposing a different architecture. High-value operational logs can remain inside Datadog for immediate workflows while high-volume or long-retention data can move into ClickHouse. Engineers still investigate from the same Datadog interface. That changes the conversation from data reduction to data placement, and those are very different decisions.
The Bigger Signal Behind the Partnership
The most interesting aspect of this announcement may not be ClickHouse itself. It may be what Datadog is doing.
Just 1 day before the partnership announcement, Datadog introduced Federated Logs support across multiple storage backends, including ClickHouse, Snowflake, Databricks, and Amazon S3. Viewed through that lens, this is not simply a bilateral integration. It is part of a larger strategy.
Datadog appears to be moving toward a storage-agnostic observability model where investigation remains centralized while storage becomes increasingly flexible. That mirrors broader trends across Enterprise Observability, Cloud Infrastructure, and AI Infrastructure.
Customers want choice. They want to control economics. They want fewer architectural constraints tied to where data happens to reside. The companies that enable those outcomes are likely to gain influence as telemetry volumes continue expanding.
What This Signals
The observability market is entering a new phase. The first era focused on collecting data. The second focused on consolidating tools. The emerging phase appears focused on decoupling storage from workflow.
ClickHouse and Datadog are not the only companies pursuing that direction, but this partnership provides one of the clearest examples yet of how that future might operate. For engineering leaders, the significance extends beyond a single integration. The question increasingly becomes whether observability platforms should own data storage or simply provide the operational layer that sits above it.
That debate is still unfolding. What is becoming harder to ignore is the volume of telemetry moving through modern systems and the growing demand for architectures that allow organizations to keep more of it without losing control of cost, access, or flexibility.
Both announced capabilities remain in preview, making general availability, customer adoption, and production deployment milestones important signals to watch over the coming quarters.
Frequently Asked Questions
What is the ClickHouse and Datadog partnership?
The partnership allows organizations to route logs from Datadog Observability Pipelines into ClickHouse and search those logs through Datadog Log Explorer without re-ingesting the data.
What is federated log search?
Federated log search enables users to query logs stored in external platforms such as ClickHouse directly from Datadog while keeping the data in its original location.
Why are observability teams interested in ClickHouse?
ClickHouse is optimized for high-performance analytics and large-scale telemetry workloads, making it attractive for long-term log retention, observability analytics, and infrastructure monitoring.
How does Datadog Observability Pipelines work with ClickHouse?
Datadog Observability Pipelines can route, enrich, parse, filter, and redact logs before delivering them directly to ClickHouse.
Is the ClickHouse and Datadog integration generally available?
No. The announced capabilities are currently in preview.
What broader trend does this partnership represent?
The partnership reflects a move toward storage-agnostic observability architectures where storage, search, and investigation workflows operate independently.
Which companies are included in Datadog's broader Federated Logs strategy?
Datadog has publicly discussed support for ClickHouse, Snowflake, Databricks, and Amazon S3 as part of its federated search approach.
What is observability in software infrastructure?
Observability platforms help engineering teams monitor, troubleshoot, and analyze software systems using logs, metrics, traces, and operational telemetry. Modern observability has become increasingly important as AI systems and distributed architectures generate larger volumes of operational data.