Skip to main content

Posts

Recent posts

Clickhouse and iceberg

Is common for ClickHouse to use S3 for data storage. And while it's not as common (yet) for ClickHouse to directly integrate with Iceberg as a table format, the integration is evolving and becoming increasingly relevant in modern data architectures. Let's break down each part: 1. ClickHouse and S3 (Common and Yes): Yes, it's very common for ClickHouse to interact with and utilize S3. In cloud deployments, especially on AWS (where S3 is native), it's a highly prevalent pattern. Why is it common? Scalability and Cost-Effectiveness: S3 provides highly scalable and cost-effective object storage. ClickHouse, designed for massive datasets, benefits greatly from this. Separation of Compute and Storage: Using S3 allows for decoupling compute (ClickHouse servers) from storage. This enables independent scaling of both and is a hallmark of modern cloud architectures. Cloud-Native Architecture: For organizations adopting cloud-native approaches, leveraging cloud-native storage like...

Why Flux

Unraveling Flux: Why Your React Frontend Needs a Data Flow Blueprint Hey Back-End Engineers, let's talk about frontends. Specifically, React frontends, which you've likely heard a lot about. You might be thinking, "Frontend? Isn't that just HTML, CSS, and a bit of JavaScript glue? We handle the real data logic on the server."  Well, in today's web applications, that "bit of JavaScript glue" is doing a lot more, and that's where things like Flux come in. Imagine your backend system. You likely have a well-defined architecture.  Perhaps a layered approach, clear separation of concerns, and robust data flow.  You wouldn't just let data fly around willy-nilly, would you?  The same principle applies, and becomes even more critical, in complex React applications. This article isn't about convincing you to become frontend gurus overnight. It's about explaining why a pattern like Flux became essential in the React ecosystem, addressing the ch...

Decoding Operational Complexity: Harnessing ClickHouse & Grafana for High Cardinality Metrics

  In today's data-driven world, operational metrics are the lifeblood of any organization running complex systems. They provide crucial insights into performance, availability, and user behavior. However, as systems become more intricate and user bases grow, we often encounter the daunting challenge of high cardinality metrics . Imagine tracking metrics not just by server, but by individual container, user session, product SKU, or geographical location. Suddenly, the number of unique dimensions explodes, leading to data management headaches and performance bottlenecks. This article explores how to effectively navigate the realm of high cardinality operational metrics, focusing on the powerful combination of ClickHouse and Grafana . We'll delve into the nature of high cardinality, understand why it's a challenge, and discuss practical strategies to manage it, including cost considerations and the feasibility of using free solutions. The High Cardinality Conundrum: Why is ...