Official service account of 360 Smart Cloud, dedicated to building a high-quality, secure, highly available, convenient, and stable one‑stop cloud service platform.
This article introduces eBPF, explains its zero‑intrusion observability advantages, describes its architecture and runtime workflow, outlines development prerequisites and compilation steps, and provides concrete Go‑based examples for tracing bash commands and measuring TCP connection latency.
This article introduces Knative Eventing concepts and core components, then demonstrates three practical usage patterns—Source‑to‑Sink, Channel and Subscription, and Broker and Trigger—through step‑by‑step examples, Kubernetes manifests, and command‑line operations, highlighting event flow, filtering, and reply handling.
This article explains Knative Eventing’s role in serverless event‑driven architectures, detailing its core concepts, the generic event‑listening model, key components such as Sources, Channels, Subscriptions, Brokers, Triggers, and flow resources like Sequence and Parallel, and provides practical YAML and HTTP examples.
The article presents a technical overview of a WebRTC‑based video conferencing system that employs a single‑SFU architecture, identifies scalability and latency challenges in large‑scale and global deployments, and proposes a cascaded SFU solution with detailed signaling, buffer management, and performance evaluation demonstrating improved load balancing and extensibility.
This article details the background, planning, step‑by‑step procedures, encountered issues, and rollback strategies for upgrading a Hadoop HDFS cluster from version 2.6.0‑cdh to 3.1.2, including mixed‑deployment of DataNodes across different federations and necessary configuration changes.
This article details the product models, usage scenarios, and a series of performance and operational optimizations—including query plan health checks, space reclamation, resource isolation, cloud‑native deployment, cross‑region high availability, and unified monitoring—implemented for the TiDB service operated by 360 Zhihui Cloud since its launch in April 2023.
The article introduces Archer, a new big‑data warehouse engine built on Iceberg that adds an inverted‑index mechanism using Tantivy to provide full‑text and JSON search, storage‑compute separation, and significant performance gains over traditional Elasticsearch and Iceberg connectors.
This article explains how to build a serverless image‑processing pipeline that automatically compresses and watermarks uploaded pictures by deploying Go functions, configuring event‑bus rules, enabling OBS notifications, and linking them together, highlighting cost, scalability, and maintenance advantages over traditional architectures.
This article introduces the concept of managed Kubernetes clusters, compares them with traditional setups, details the architecture and implementation process, discusses challenges such as cross‑cluster webhook routing, certificate issuance, container runtime integration, and outlines the diverse features of the dedicated cluster solution.
The article outlines the challenges of massive AI training data, defines storage performance requirements, and presents Polefs—a cloud‑native distributed cache file system with unified storage, metadata acceleration, and read/write caching designed to improve GPU utilization and reduce data redundancy.
