This article explains eight common software architecture patterns—from single‑database apps to microservices, caching, sharding, elastic scaling and multi‑datacenter deployment—detailing their designs, typical use cases, advantages, drawbacks, and practical implementation steps.
This guide explains the two‑location three‑center disaster‑recovery pattern, describing its purpose, typical deployment across two cities and three data centers, and step‑by‑step recommendations for same‑city dual‑active or primary‑backup setups, remote backup strategies, traffic routing, and essential monitoring.
This article explains TiDB's distributed, financial‑grade high‑availability architecture, covering single‑cluster same‑zone multi‑datacenter deployment, cross‑cluster DTS synchronization, underlying Raft and label mechanisms, configuration examples, performance trade‑offs, and real‑world monitoring results on the HULK cloud platform.
Learn how to execute a zero‑downtime, risk‑controlled database migration across data centers using pre‑expansion, cross‑room master switch, intelligent domain scheduling, and step‑by‑step operational guides—including VIP handling, global vs. zone‑specific domains, and post‑migration validation—to ensure continuous service and optimal resource elasticity.
This article presents an in‑depth overview of ByteDance's EB‑scale HDFS, covering its new features, multi‑datacenter architecture, tiered storage implementation, data management services, capacity and fault‑tolerance strategies, as well as practical data‑protection mechanisms and related Q&A.
This article details Bilibili's multi‑datacenter offline architecture, explaining the capacity challenges, the chosen scale‑out design, and the implementation of job placement, data replication, routing, versioning, throttling, and traffic analysis to efficiently handle massive batch workloads across geographically distributed clusters.
This article describes Bilibili's offline multi‑datacenter architecture, explaining why a scale‑out approach was chosen over scale‑up, and detailing the unit‑based design, job placement, data replication, routing, versioning, bandwidth throttling, traffic analysis, and the operational results and future directions.
This article details Bilibili's multi‑datacenter solution for offline big‑data workloads, covering the challenges of capacity limits, the design of a unit‑based architecture, job placement, data replication, routing, versioning, bandwidth throttling, traffic analysis, and future directions.
To overcome rapid data growth and on‑premise capacity limits, Bilibili adopted a scale‑out, unit‑based multi‑datacenter architecture that isolates failures, intelligently places jobs, replicates data via an enhanced DistCp service, routes reads with an IP‑aware HDFS router, and throttles cross‑site traffic, enabling stable offline big‑data processing of hundreds of petabytes while preserving throughput.
This article explains how ByteDance extended the Apache HDFS architecture with a multi‑datacenter design, introducing components such as DanceNN, NNProxy, and BookKeeper to achieve scalable storage, cross‑datacenter data placement, and rack‑level disaster recovery for petabyte‑scale workloads.
This article explains how a company extended native Redis to support bidirectional synchronization across multiple data‑center sites, addressing issues such as lack of master‑master replication, data loops, idempotency, write conflicts, and providing a custom rLog design for efficient breakpoint‑resume and performance.
The article chronicles a YouZan DevOps engineer’s five‑year evolution from Alibaba‑based middleware duties to business‑operation leadership, highlighting the relentless pursuit of system stability through the 1‑minute detection, 5‑minute localization, 10‑minute resolution mantra, complex multi‑datacenter integrations, continuous learning, and a mindset of proactive problem‑solving.
This article walks through the evolution of a Redis‑based cache layer for multi‑data‑center deployments, addressing consistency, safety, performance, disk‑space, data loops, timestamp versioning, master‑slave failover, and global numeric aggregation, and culminates in a ready‑to‑use middleware solution.
This article explains the design and implementation of NSQ dual‑ and multi‑data‑center architectures using a lookup‑migrate proxy, covering deployment scenarios, routing strategies, migration phases, JSON response transformations, and practical lessons learned for reliable message publishing and consumption across data centers.
This article examines Alipay's evolution from a simple single‑datacenter setup to a multi‑active‑active, unit‑based architecture, detailing the technical challenges of high availability, disaster recovery, failover design, blue‑green deployment, and how these solutions enable continuous service during massive traffic spikes like Double 11.
This article examines the importance of high‑availability and disaster‑recovery architectures, tracing Alipay’s evolution from a simple load‑balanced setup through multi‑datacenter, failover, and unit‑based designs that address scalability, data consistency, and continuous service delivery challenges.
This article outlines a comprehensive Redis multi‑datacenter deployment plan, detailing master‑backup placement, tree‑structured master‑slave topology across three sites, configuration checks, Pull&Push synchronization mechanics, and includes concrete code snippets illustrating connection, RDB generation, and incremental replication processes.
After initially focusing on a distributed MySQL system, the author describes why open‑source HA solutions like MHA were unsuitable, then details the design and implementation of a 4‑layer NAT‑based proxy (RDS) and a more advanced 7‑layer application‑level proxy, highlighting features such as authentication, load balancing, read/write splitting, and multi‑datacenter awareness.
The article explains how multi‑datacenter architectures prevent total service loss, improve latency by placing services near users, and balance the CAP trade‑offs through models like AC, CP, and AP, while outlining practical design, sharding, monitoring, and failover strategies for large‑scale backend systems.
This article details Meizu's cloud synchronization service, covering its custom MZ‑SyncML protocol, semi‑sync mechanisms, file and one‑sync protocols, failure handling, service architecture, modular design, massive data routing, multi‑datacenter deployment, traffic optimization, and practical lessons learned.
Meizu moved from a single‑datacenter to a multi‑datacenter architecture to improve reliability, reduce latency, and meet user proximity demands, detailing technical challenges, traffic scheduling, read‑heavy and read‑write balanced services, and GSLB‑based routing solutions.
