This guide outlines practical strategies for designing and optimizing network and storage in Kubernetes clusters of over 5,000 nodes, covering overlay networks, IP pool segmentation, bandwidth allocation, load balancing, security policies, distributed storage options, performance tuning, and reliable backup solutions.
This article explains how to design the storage layer for an e‑commerce shopping cart, covering core functionalities, required attributes, client‑side temporary storage, server‑side persistence options with MySQL and Redis, and the trade‑offs between reliability, performance, and complexity.
This article explains the evolution of instant messaging systems from traditional to modern architectures, detailing message synchronization and storage mechanisms, inbox/outbox models, read/write diffusion strategies, and the practical implementation using Redis, MongoDB, and MySQL to achieve scalable, reliable communication.
This article explains the core storage design of RocketMQ, covering the CommitLog, ConsumeQueue, and Index files, their organization, sequential write strategy, memory‑mapped I/O, and flexible flushing policies that together provide high‑throughput, low‑latency messaging for backend systems.
This article explains how to design a configuration center, covering storage selection, environment isolation, push‑pull update mechanisms, Spring integration, permission auditing, metrics monitoring, and open APIs or multi‑language SDKs for practical implementation.
This article provides a comprehensive overview of Apache RocketMQ’s core architecture, including topic routing mechanisms, message storage file designs, high‑availability message sending, concurrent pull and consumption processes, HA synchronization, and transaction messaging, while offering practical learning steps and programming techniques for developers.
The article presents JD's self‑developed JMQ middleware, detailing its 2018 performance improvements, comparative benchmarks with Kafka, storage and index architecture evolution from JMQ2 to JMQ4, and the high‑performance I/O and caching strategies that enable millions of TPS on modern hardware.
The article presents an in‑depth technical overview of JD's self‑developed JMQ middleware, detailing its performance improvements over previous versions, comparing write throughput with Kafka under synchronous and asynchronous scenarios, and explaining JMQ4's storage architecture, dense indexing, high‑performance I/O, and cache management strategies.
