ClickStack's January Surge: Faster Observability
Alps Wang
Feb 12, 2026 · 1 views
Unlocking Observability's Potential
This article highlights significant advancements in ClickStack, particularly around index support and the introduction of Managed ClickStack. The shift to a more flexible and performant indexing strategy, especially with the adoption of bloom_filter and text indices, is a welcome improvement. The detailed examples and performance comparisons provided in the article are valuable for developers seeking to optimize their observability pipelines. The move to Managed ClickStack, with its compute and storage-based pricing model, is a significant change in the observability landscape, potentially lowering the barrier to entry for teams seeking high-performance log, metric, and trace analysis. The integration of materialized views for alerts is also a clever optimization, improving efficiency and scalability.
However, there are a few areas that warrant further consideration. The article emphasizes the importance of tuning, but the complexity of index tuning and the potential for performance degradation if not done correctly is a risk. While the move to bloom_filter reduces tuning parameters, the performance gains are highly dependent on data characteristics. The reliance on the splitByNonAlpha tokenizer for the text index might limit its applicability for certain data types and use cases. The constraints around the maximum number of tokens per call for hasAllTokens and the dependency on ClickHouse version 25.12 are potential limitations. Finally, while the article provides a good overview, it would benefit from more detailed benchmarking data across a broader range of datasets and workloads to fully quantify the impact of these changes. More explicit guidance on when to choose inverted indices vs. Bloom filters would be beneficial for users.
Overall, the January updates to ClickStack represent a substantial leap forward. The improvements in index support, the introduction of Managed ClickStack, and the integration of materialized views are all positive developments. However, users should carefully consider the tuning requirements and limitations when implementing these features to ensure optimal performance and avoid potential pitfalls. Continued benchmarking and more detailed guidance from ClickHouse will be crucial for maximizing the benefits of these new capabilities.
Key Points
- Managed ClickStack is a new fully managed observability offering on ClickHouse Cloud, priced on compute and storage, not events, hosts, or users.
- Significant improvements in index support: the simpler
bloom_filterindex and the introduction of inverted text indexes, improving query performance and reducing tuning complexity. - Alerts now leverage materialized views, making alert evaluations more lightweight and efficient, increasing the number of alerts that can be run without impacting performance.

📖 Source: What's new in ClickStack. January '26.
Related Articles
Comments (0)
No comments yet. Be the first to comment!
