This article presents an end‑to‑end design of a production‑grade observability platform that ingests millions of real‑time logs, metrics, and events, detailing functional and non‑functional requirements, capacity planning, component choices such as Kafka, Flink, Elasticsearch, object‑storage data lakes, and the trade‑offs involved.
This article walks through the challenges of scaling Prometheus storage, compares Thanos, Cortex, and VictoriaMetrics, and provides a complete step‑by‑step guide—including hardware requirements, configuration, deployment, tuning, multi‑tenant setup, and troubleshooting—to replace Prometheus local TSDB with VictoriaMetrics for long‑term, high‑performance monitoring.
In today's cloud‑native era, Prometheus and Grafana have become the de‑facto standard for full‑stack monitoring across servers, databases, containers, and applications, offering multi‑dimensional data models, flexible queries, and alerting that are essential skills for developers, SREs, and ops engineers.
This article presents a detailed technical case study of using KaiwuDB’s multi‑model database to unify card‑based and coin‑based payment processing in amusement parks, covering architecture, schema design, SQL implementations, offline handling, cross‑model analytics, hot‑cold data tiering, visualization, monitoring, security, and high‑availability strategies.
Grafana Mimir 3.0 introduces a decoupled read‑write architecture, a streaming query engine that cuts memory use by up to 92%, and cost‑saving optimizations that reduce resource usage by 15%, while providing detailed upgrade guidance for large‑scale TSDB deployments.
The article profiles Tianmou Technology’s co‑founder and CTO Qiao Jialin, detailing his personal journey, the company’s mission to build the world’s best time‑series database, its technical advantages, adoption by aerospace and other industries, and the cultural values that sustain rapid innovation.
The article summarizes the 2025 Time Series Database Innovation Conference where Apache IoTDB’s evolution, technical breakthroughs, benchmark leadership, open‑source community growth, and real‑world industrial deployments from aerospace to oil‑gas are detailed, highlighting the upcoming IoTDB 2.0 vision.
VictoriaMetrics employs a modular, horizontally scalable architecture composed of vmagent, vminsert, vmstorage, vmselect, and vmalert, each handling data collection, ingestion, storage, querying, and alerting, while leveraging consistent hashing, LSM‑tree storage, TSID indexing, and multi‑tenant isolation to efficiently manage large‑scale time‑series workloads.
This article explains the challenges of managing heterogeneous IoT sensor data, compares wide‑table and narrow‑table modeling approaches, and introduces DolphinDB's IOTDB engine—featuring a unified IOTANY column, a TSDB engine, a latest‑value cache, and a static info table—to efficiently store and query diverse point data in a single table.
Apache HoraeDB (CeresDB) is an open‑source, distributed, high‑availability time‑series database developed by Ant Group, supporting multi‑dimensional queries, compatible with Prometheus and OpenTSDB, and offering SQL and OLAP capabilities for use cases such as APM, IoT monitoring, financial analytics, and AI‑infra observability.
This article examines the migration from Thanos to VictoriaMetrics for large‑scale metric storage, detailing background challenges, VictoriaMetrics architecture and storage engine, data write and read processes, and a comparative analysis of performance, scalability, and operational costs between the two systems.
The author reflects on a decade‑plus journey through China’s DTCC database conferences, describing personal growth from novice to speaker and organizer, sharing insights on Redis Cluster, distributed database selection, openGauss, time‑series databases, and the evolving themes that chart the industry's progress.
Qunar’s Watcher monitoring platform was upgraded from minute‑level to second‑level precision, redesigning storage, data collection, and alerting pipelines, adopting VictoriaMetrics, enhancing client SDKs, and adding fine‑grained alarm rules, which reduced fault detection from four minutes to under one minute while improving reliability and scalability.
This article summarizes a live session where Shawn explains how understanding business scenarios drives effective OpenGemini database design, and provides comprehensive best‑practice guidance on library and table design, data ingestion, query optimization, and performance tuning for time‑series workloads.
This article explains why InfluxDB stands out among time‑series databases by covering its performance, storage model, use cases, comparison with relational databases, line protocol fundamentals, Docker installation steps, and PHP client integration, providing a practical guide for developers.
Cloud Music’s middleware team built the Pylon APM monitoring system on VictoriaMetrics, combining exporters, vmagent, Nacos, Flink‑based pre‑aggregation recording rules and vminsert for collection with Grafana, a custom Proxy and vmselect for querying, achieving millisecond‑level latency, metric‑trace correlation, stability improvements, and cost‑effective storage for nearly 700 million active time series.
HoraeDB, the open‑source time‑series database formerly known as CeresDB, has been accepted into the Apache Incubator, and the announcement details its design objectives, cloud‑native architecture, distributed solution, key features such as high performance, low cost, compute‑storage separation, and how developers can contribute to the project.
Facing exponential growth of time‑series data from 2017 to 2023, Didi migrated from InfluxDB to RRDtool, then to an in‑memory cache layer, and finally adopted VictoriaMetrics because its low‑cost commodity‑hardware operation, high write throughput, strong compression, and easy horizontal scaling solved the earlier storage, OOM, and scalability problems.
This article explains Ant Group's client observability system, the technical difficulties of tracing, logging, and metrics on mobile clients, and presents their open‑source solutions—including a custom time‑series database, dimension‑join services, and intelligent alerting—to handle massive data and multi‑dimensional analysis.
This article compares three leading time‑series databases—InfluxDB, Kdb+, and Prometheus—detailing their origins, core features, strengths, and drawbacks, and helps readers decide which solution best fits specific monitoring, IoT, or financial data workloads.
This guide explains InfluxDB's architecture, data model, CRUD commands, measurement handling, retention policies, query syntax, time‑zone considerations, and service management, providing a comprehensive tutorial for developers working with time‑series databases.
An in‑depth comparison of Prometheus and Zabbix examines their histories, architectures, data storage, scalability, and container support, highlighting Prometheus’s cloud‑native pull model and Go‑based performance versus Zabbix’s mature, relational‑database approach, to help teams choose the right monitoring solution.
This article details Qunar's migration from a Prometheus‑based monitoring stack to VictoriaMetrics, describing the limitations they faced, the architectural redesign using vmagent, vmcluster, and vmalert, and the resulting performance improvements and operational benefits for large‑scale Kubernetes environments.
This article details Qunar’s migration from a traditional Prometheus‑based monitoring stack to VictoriaMetrics, describing the challenges of large‑scale container metrics collection, the architectural redesign using VM‑Cluster, vmagent, and vmalert, and the performance improvements achieved after full replacement.
This article details Ctrip’s large‑scale log monitoring architecture, covering the overall Overview, the Clog log system, the CAT tracing platform, and the internal TSDB solution, explaining how billions of logs are processed in real time with low latency, high reliability, and efficient querying.
This article examines the data challenges faced by intelligent operations (AIOps), evaluates IoTDB against other time‑series databases through performance benchmarks, outlines Cloudwise's architecture and open‑source contributions, and presents real‑world case studies demonstrating anomaly detection and root‑cause analysis in industrial settings.
This article details the evolution of NetEase's self‑built time‑series database EyesTSDB into a cloud‑native, second‑level monitoring solution, covering its architecture, core features, integration with VictoriaMetrics, custom plugin workflow, CMDB linkage, real‑world use cases, and future challenges.
The TDengine Developer Conference in Beijing unveiled the open‑source, cloud‑native TDengine 3.0, detailing its revolutionary architecture that tackles high‑cardinality challenges, introduces RAFT‑based distribution, and showcases real‑world IoT and IT‑operations case studies where enterprises dramatically improved performance and reduced costs.
CeresDB, a distributed high‑availability time‑series database originally built at Ant Group, is now open‑sourced with version 0.2.0, offering high‑throughput writes, multi‑dimensional queries, SQL support, compatibility with Prometheus and OpenTSDB, and a range of features targeting both monitoring and analytical workloads.
This article introduces Apache IoTDB, an open‑file based time‑series database designed for industrial IoT, explains its TsFile storage format, data modeling options, layered architecture (embedded, edge, cloud), performance advantages over traditional formats, and highlights the active open‑source community and real‑world deployments.
This article introduces Apache IoTDB, an open‑source time‑series database for industrial IoT, outlines its recent feature releases, explains its data‑modeling and compression strategies, and discusses UDF, trigger, and quality‑control capabilities that guide technical selection and architecture design.
Vivo’s vmonitor system replaces its legacy RabbitMQ‑based pipeline with an HTTP‑driven collector and gateway, stores minute‑level JVM, system, and business metrics in a customized OpenTSDB on HBase, adds precise floating‑point handling and null‑aware aggregation, buffers data in Redis, and provides multi‑dimensional alerts comparable to Zabbix, Open‑Falcon, and Prometheus.
This article explains the fundamentals of time‑series data, outlines the architecture and core technical advantages of Baidu Cloud's TSDB, and demonstrates how the database powers IoT, smart‑city, industrial, power‑grid, and autonomous‑driving use cases through multi‑level storage, distributed query optimization, and edge‑cloud integration.
This article introduces VictoriaMetrics, a fast and cost‑effective time‑series monitoring solution that can replace Prometheus, covering its definition, key features, core components, and essential installation considerations for both binary and Kubernetes deployments.
This article explains how Prometheus’s time‑series database handles massive monitoring data, illustrates practical query examples, and shows why its storage engine and pre‑computation features enable efficient, high‑performance observability for large‑scale services.
This article explains how Prometheus evolved from a Google internal project to a CNCF‑graduated, top‑ranked time‑series database and full‑stack monitoring ecosystem, detailing its history, core features, architecture, and the roles of its components such as Exporters, Pushgateway, Service Discovery, and Alertmanager.
This article explains what time series databases are, introduces InfluxDB as the leading TSDB, describes its TICK architecture and storage engine, provides step‑by‑step installation and configuration of InfluxDB and Telegraf, demonstrates visualization, JMeter integration, and Flux queries in Python, and highlights the rapid market growth of TSDBs.
Learn how to configure and use the open‑source time‑series database InfluxDB in a Spring Boot application, covering essential concepts, dependency setup, property configuration, scheduled data writing, and verification via InfluxDB CLI, with complete code examples.
This article explores the evolution and current landscape of time‑series databases, detailing design principles, storage structures such as B‑Tree, B+Tree, and LSM‑Tree, and providing an in‑depth comparison of OpenTSDB, InfluxDB, and the emerging Apache IoTDB, while also discussing practical deployment considerations and industry use cases.
This article reviews the rapid growth of global data, examines the evolving landscape and classification of time‑series databases, analyzes storage engine designs such as B‑Tree versus LSM‑Tree, discusses query optimization and real‑time analytics, and outlines practical application scenarios in IoT and industrial settings.
DeWu chose the open‑source time‑series database TDengine to store billions of daily Sentinel flow‑control metrics, using a super‑table design with per‑resource tables, a three‑node cluster, Druid/MyBatis pooling, and raw‑SQL writes, achieving 10 ms batch write latency, sub‑millisecond queries, and 90 % compression.
This article introduces Apache IoTDB, a high‑performance time‑series database for IoT, covering its key features, overall architecture, installation procedures, command‑line operations, and JDBC usage with example code to help newcomers quickly start storing and querying sensor data.
This article explains the fundamentals, application scenarios, key requirements, and open‑source options for time series databases, highlighting how GaussDB (For Influx) addresses high‑performance writes, massive timelines, low storage cost, and elastic scaling for IoT and cloud monitoring workloads.
This article explains how Prometheus uses a time‑series database (TSDB) to handle massive monitoring data, detailing its concepts, query examples, storage engine design, indexing mechanisms, and the benefits of pre‑computing expressions for efficient real‑time analysis.
This article introduces JD Cloud’s self‑built time‑series database HoraeDB, explaining its core concepts, typical use cases, architectural layers, high‑performance features, down‑sampling strategies, compression techniques, and stability measures for handling massive, 24‑hour monitoring data at scale.
The article offers a detailed technical overview of OpenTSDB’s architecture and data model, explaining how it leverages HBase for scalable time‑series storage, describing core concepts, table schemas, ingestion flow, performance considerations, and future alternatives for large‑scale monitoring workloads.
Prometheus, a CNCF‑graduated, cloud‑native monitoring system, combines pull‑based target discovery, a label‑rich time‑series data model, and four core metric types—gauge, counter, histogram, and summary—to provide near‑real‑time visibility, short‑term retention, alerting via AlertManager, and integration with Grafana and remote storage for scalable observability.
The article details 360's external network quality monitoring system, explaining its background, real‑time detection features, CDN‑based source host selection, three‑layer architecture, data collection and storage pipelines, fault‑diagnosis strategies, and visualization approaches for rapid network fault localization.
The article describes how 360 implements an external network quality monitoring system that uses CDN nodes as source hosts to perform minute‑level, end‑to‑end ping measurements, stores results in time‑series and other databases, analyzes them to detect VIP, data‑center or ISP faults, and generates visualized alerts and reports for operations teams.
Jiangsu Mobile’s IT operations team partnered with Newland to build a high‑availability, real‑time performance management platform using Prometheus, achieving billion‑level call‑record monitoring, low‑latency queries, data compression, and advanced forecasting, dramatically improving system health visibility and operational efficiency.
To mitigate the inherent single‑point‑failure risk of the open‑source InfluxDB community edition, the article proposes deploying multiple InfluxDB instances with concurrent client writes, tracking failed writes, temporarily storing them, and using custom workers to replay data, while addressing timeout, data consistency, and storage considerations.
This article examines the challenges of selecting an IoT big‑data platform, compares traditional real‑time databases, Hadoop‑based solutions, and modern time‑series databases such as TDengine, InfluxDB and ClickHouse, and provides practical case studies and criteria for making an informed choice.
This article explains InfluxDB's storage engine workflow—including WAL, Cache, TSM files, compression components, and file management—then provides hardware sizing guidance based on write/query load, series cardinality, and recommends SSD storage with sample configuration settings.
This article explains the fundamentals of time‑series databases, introduces InfluxDB’s key components such as databases, measurements, retention policies, tags, fields, series, and points, describes the line protocol syntax, compares it with traditional databases, and outlines the design trade‑offs that shape its performance and limitations.
This article introduces Prometheus, compares it with Zabbix, Open‑Falcon and Nagios, explains its architecture, data model, exporters, storage options, query language, alerting and federation, and shares practical deployment experiences and common Q&A for cloud‑native environments.
QTSDB is a distributed time‑series database built on InfluxDB 1.7 that adds cluster capabilities such as horizontal scaling, replica fault‑tolerance, automatic data expiration, and high‑throughput writes, while preserving InfluxDB's query syntax and storage engine features.
This article introduces OpenTSDB as a distributed, scalable time‑series database built on HBase, explains its architecture, data model, and storage optimizations, presents real‑world monitoring use cases, analyzes performance issues caused by high‑cardinality tags, and details the solution steps taken to restore query speed.
This article presents a comprehensive overview of Alibaba's Time Series Database (TSDB), covering its business panorama, covered scenarios, challenges, performance evolution, architecture, core compression and query technologies, stability mechanisms, and future development plans.
This article presents the design, architecture, and technology choices of a Sonar monitoring system for Spring Boot microservices, covering time‑series database selection, deployment topology, client collection strategies, and future plans for advanced analytics and AI‑driven alerting.
Baidu’s TSDB processes over 20 million data points per second per node and tens of thousands of queries per second cluster‑wide by employing a stateless read/write‑separated elastic architecture, multi‑layer storage across Redis, HBase and Hadoop, minute‑level geo‑redundant self‑healing, and a modified Gorilla compression that cuts storage by 80% with minimal CPU overhead.
This article examines the architectural redesign of Alibaba's High‑Performance Time Series Database (HiTSDB), covering storage model changes, inverted‑index enhancements, a pipelined streaming aggregation engine, data‑migration strategies, and performance benchmarks that together deliver over tenfold query speed improvements.
This article introduces the concept, characteristics, and use cases of time series databases, explains the data model and challenges of traditional solutions, and provides a detailed overview of Tencent's Cloud Time Series Database (CTSDB) along with performance comparisons against InfluxDB.
This article introduces the fundamentals of time‑series databases, compares them with traditional databases, surveys industry adoption, and details how Tiger Live leverages OpenTSDB, Grafana, and Bosun to build a scalable monitoring system for live‑streaming services.
This article explains the fundamentals, use cases, challenges, and storage architectures of time‑series databases, illustrating how they handle massive IoT telemetry through log‑structured storage, sharding, and distributed designs while supporting high‑throughput writes and fast analytical queries.
This article translates and expands a comparative analysis of Prometheus against Graphite, InfluxDB, OpenTSDB, Nagios, and Sensu, covering their applicability, data models, storage mechanisms, architecture, and strengths to help readers choose the most suitable time‑series database for monitoring needs.
This article presents a comprehensive case study of a business monitoring system, covering its background, architectural analysis, module design, time‑series database selection, visualization with Grafana, alerting strategies, decision‑making logic, and intelligent monitoring experiments, followed by key takeaways and lessons learned.
