This article breaks down the four-step architecture design of a high-concurrency encyclopedia system, covering latency-driven multi-data-center deployment, a five-layer request chain with caching at each level, a three-tier optimization pyramid, and master-slave synchronization with automatic failover for disaster recovery.
This article details Vivo's approach to managing thousands of Kubernetes nodes across multiple data centers by developing a declarative Kubernetes‑Operator, modular Ansible scripts, and a comprehensive CI matrix to automate deployment, scaling, upgrades, and fault recovery while reducing operational risk.
Vivo replaced error‑prone manual Ansible playbooks with a custom Kubernetes Operator that uses declarative CRDs and modular Ansible scripts to automate the full lifecycle—deployment, scaling, upgrades, and recovery—of thousands of nodes across multiple data‑centers, supported by extensive CI testing and future kubeadm integration.
This article describes the challenges of native Redis in multi‑data‑center deployments and presents the design, implementation, and performance evaluation of a custom Redis extension that adds bidirectional synchronization, rlog logging, protocol enhancements, and conflict‑resolution mechanisms to achieve reliable cross‑region active‑active operation.
Suning built a multi‑data‑center active‑active solution that combines primary‑backup, same‑city active‑active, and full multi‑active modes, defines top‑level design goals, values and principles, and implements a comprehensive architecture, routing, high‑availability, hybrid‑cloud and disaster‑recovery strategy to support massive e‑commerce growth.
This article explains how MongoDB's nearest mode achieves proximity‑aware reads across multiple data centers by analyzing the internal mongos and driver code, detailing latency collection, smoothing algorithms, node selection logic, and providing configuration recommendations for latency‑sensitive workloads.
This article examines the requirements of active‑active applications across multiple data centers, compares distributed database architectures—including two‑phase commit, multi‑master, and sharded designs—and explains how MongoDB’s sharding and write/read concerns enable high availability, consistency, and rapid failover in such environments.
The article explains eBay's use of MongoDB to create highly available, fault‑tolerant multi‑data‑center architectures, detailing design patterns, replica set configurations, read/write strategies, and recent MongoDB features that enable scalable, mission‑critical applications.
After Meituan merged with Dianping, engineers unified two massive Hadoop ecosystems across Beijing and Shanghai by breaking the project into four phases—unify, copy, switch, fuse—standardizing versions, implementing zone‑aware transfers, cross‑realm Kerberos, and federated metadata to achieve a single, reliable multi‑data‑center platform.
This article introduces XPipe, a framework designed by Ctrip to enable Redis multi‑data‑center replication, ensure high availability through keeper and MetaServer components, and provide reliable disaster‑recovery (DR) switching while maintaining low latency and data consistency.
The article details Weibo's journey in building a multi‑data‑center active‑active architecture, covering its evolution, technical challenges such as latency and data synchronization, the adopted MCQ‑based messaging solution, operational best practices, and future directions for high‑availability deployments.
This article examines the characteristics and billing models of multi‑data‑center networks, analyzes external traffic patterns, identifies challenges in optimizing Internet‑facing bandwidth, and proposes practical scheduling strategies to better utilize idle bandwidth and reduce carrier costs.
