Prometheus Programming Tutorials & Engineering Articles

AI infrastructure spans multiple layers, from compute and networking to storage, orchestration, and applications. When performance degrades…

Explore ClickStack’s latest upgrades, from richer trace navigation and Prometheus connectivity to smarter dashboards, quieter alerts, faster filtering and support for exponential histogram metrics.

How we rebuilt ClickHouse Cloud's autoscaling orchestration on Kubernetes' controller-runtime and a ClickHouse-powered signals table, adding a reactive fast path that scales services up in seconds instead of waiting for the next scheduled pass.

Using ClickHouse for observability and wondering which interface best fits your workflows? We explore ClickStack, Grafana, and when it makes sense to use both.

Deploying an AI coding assistant in a regulated, sovereign, or source-sensitive environment, often comes with challenges. Three common issues are: the source…

Februarys' ClickStack update brings major improvements to both Cloud and open source, including new query workflows, enhanced metrics exploration, performance optimizations, and expanded alerting options.

Retention limits, sampling, and metric roll-ups aren't observability best practices - they're workarounds for storage systems that can't handle full-fidelity data, and they're becoming a hard blocker for AI-driven workflows.

Why does high cardinality break Prometheus but not ClickHouse? In Part 1, we explore the architectural tradeoffs of Prometheus and other series-based systems, showing how cardinality impacts memory, ingestion, querying, and operational stability at scale.

In this blog, we explore why high cardinality behaves fundamentally differently in column-oriented databases like ClickHouse, and why the costs appear in very different places than in traditional series-based systems like Prometheus.

On June 15, 2016, ClickHouse went open source under the Apache 2.0 license. Ten years later, it's one of the most widely deployed analytical databases in the world.

We’ve shipped dashboard actions, custom number palettes, flexible event patterns, a built-in RUM experience, major trace view improvements, expanded MCP capabilities, scoped dashboard filters, and much more in ClickStack this May.

Maximizing the value of AI infrastructure demands deep visibility into GPU utilization. Yet many platform teams running AI workloads on Kubernetes operate with…

Distributed deep learning depends on fast, reliable GPU-to-GPU communication using the NVIDIA Collective Communication Library (NCCL). When training slows down…

Designing monitoring that works when everything else doesn’t.

How we built a storage system that ingests 50 million samples per second and stores 2.5 petabytes of logical time series data.

Slurm is an open source cluster management and job scheduling system for Linux. It manages job scheduling for over 65% of TOP500 systems.

A production-tested approach for moving a large-scale metrics pipeline from StatsD to OpenTelemetry and Prometheus.

In production Kubernetes environments, the difference between model requirements and GPU size creates inefficiencies. Lightweight automatic speech recognition…

As large language model (LLM) inference workloads grow in complexity, a single monolithic serving process starts to hit its limits. Prefill and decode stages…

How a complex, large-scale migration to an in-house observability platform led to superior tooling, consistent data, and a fundamental…

How we changed our Observability as Code alert review process and cut development cycles from weeks to minutes.

The article discusses the evolving role of metrics in observability, emphasizing that while they remain important, their function is shifting towards being an optimization layer rather than the cor...

We’re sharing details of the role backend aggregation (BAG) plays in building Meta’s gigawatt-scale AI clusters like Prometheus. BAG allows us to seamlessly connect thousands of GPUs across multipl…

The article discusses the introduction of time-based fairshare in NVIDIA Run:ai v2.

The article discusses the NVIDIA Multi-Agent Intelligent Warehouse (MAIW), an AI command layer designed to enhance operational efficiency and supply chain intelligence in automated warehouses.

Uber Engineering details their migration from a legacy monolithic monitoring system to a modern, cloud-native observability platform for their corporate network infrastructure.

The article reviews the significant developments and features introduced in ClickStack over its first seven months since launch, highlighting advancements such as JSON support, integration with Cli...

This article discusses the implementation of horizontal autoscaling for Retrieval-Augmented Generation (RAG) components on Kubernetes, focusing on NVIDIA's microservices architecture.

The article discusses the deployment of secure, data-driven AI agents using NVIDIA's AI-Q Research Assistant and Enterprise RAG Blueprints on AWS.

The article discusses the challenges of identifying the root cause of configuration management failures using Salt at Cloudflare, particularly when dealing with a high volume of changes across nume...

The article discusses the advancements presented at the Open Compute Project (OCP) Summit 2025, focusing on the evolution of networking hardware for AI applications.

The article announces that the Cadence project has joined the Cloud Native Computing Foundation (CNCF), highlighting its commitment to open-source development.

The article discusses Meta's evolution in infrastructure over 21 years, highlighting the significant changes brought about by AI.

The article discusses the critical role of networking in supporting AI infrastructure, highlighting insights from the @Scale: Networking 2025 event where industry leaders shared advancements in AI ...

The article discusses how NVIDIA Dynamo can help reduce Key-Value (KV) Cache bottlenecks in large language model (LLM) inference by offloading cache data to more cost-effective storage solutions.

The article discusses the rising costs associated with observability in software engineering and proposes a shift towards open, cost-efficient architectures.

The article discusses Pinterest's transition to Moka, a next-generation data processing platform built on AWS Elastic Kubernetes Service (EKS).

ClickHouse version 25. 8 introduces 45 new features, 47 performance optimizations, and 119 bug fixes, enhancing its capabilities as a high-performance analytical database.

Dynamo 0. 4 introduces significant enhancements for deploying large language models (LLMs) with a focus on performance, observability, and autoscaling based on service-level objectives (SLO).

This article discusses Pinterest's transition from a Hadoop-based platform to a Kubernetes-based data processing solution named Moka.

The article discusses the importance of LLM observability using ClickStack, OpenTelemetry, and MCP, highlighting how to instrument LibreChat for enhanced insights into AI-driven applications.

The article discusses the evolution of ClickHouse's observability platform, LogHouse, as it scales beyond 100 petabytes of data.

This article discusses the challenges of extracting insights from multimodal documents and presents a solution using the NVIDIA NeMo Retriever extraction pipeline.