Official service account of 360 Smart Cloud, dedicated to building a high-quality, secure, highly available, convenient, and stable one‑stop cloud service platform.
DGL Operator is an open‑source Kubernetes controller that automates the lifecycle of distributed graph neural network training by handling configuration generation, graph partitioning, training execution, and resource cleanup, providing a cloud‑native solution for large‑scale GNN workloads.
This article explains how Linux manages physical address space using SMP and NUMA architectures, describes the node, zone, and page data structures, details page allocation via the buddy system and slab allocator, and outlines user‑ and kernel‑mode page‑fault handling, swapping, and address translation mechanisms.
This article explains how Linux manages virtual memory by describing the organization of virtual and physical address spaces, the mechanisms of segmentation and paging, multi‑level page tables, and the layout of user‑mode and kernel‑mode memory, including relevant kernel code examples.
This article explains how EMQX 3.0 maintains routing and distribution layers, detailing the subscriber tables, shared‑subscription handling, topic‑trie structures, route look‑ups, and the Erlang code that drives message matching and dispatch across a clustered broker.
This article explains what rate limiting is, why it is needed, the typical actions taken when limits are reached, compares single‑node and distributed architectures, reviews four classic limiting algorithms, and provides a practical Redis‑Lua token‑bucket implementation with code examples.
This article introduces the open‑source EMQX MQTT broker, explains the MQTT protocol basics, and provides an in‑depth analysis of the broker's connection and session processes—including their state structures, message flows for connect, subscribe, unsubscribe, ping, disconnect, and QoS 0/1/2 publishing—supported by illustrative code snippets and diagrams.
This article explains why and when to scan product code for vulnerabilities, describes static source‑code and binary scanning methods, introduces deep code‑search techniques and a real‑time Sphinx‑based indexing architecture, and shows how these practices can significantly raise overall product quality.
This article explains the key differences between consortium (permissioned) and public blockchains, details Hyperledger Fabric’s architecture—including MSP, channels, and ordering services—and describes how Fabric implements an admission mechanism to control network membership.
This article introduces NSQ, an open‑source Go‑based distributed message queue, detailing its core components, scalable high‑availability deployment topology, backup and replay mechanisms, and monitoring practices to ensure reliable, fault‑tolerant message processing.
This article shares practical experience with TiDB Data Migration (DM), covering its background, architecture, key features, online DDL support, common error handling such as duplicate‑key issues, large‑scale import tuning, configuration limits, and cleanup recommendations for reliable MySQL‑to‑TiDB synchronization.
