Search across authors, categories, and technical themes. The layout mirrors the editorial references while staying responsive and fast.
During the 24th GOPS Global Operations Conference, Migu Interactive Entertainment’s Vice President Su Yi discussed how their AI‑driven AIOps observability framework, validated by ITU standards, enhances cloud gaming platform stability, accelerates issue detection, and supports China Mobile’s 5G‑based digital transformation.
This article examines how test teams can enhance application observability and quality assurance by distinguishing observability from traditional monitoring, defining goals, outlining a monitoring foundation, and proposing module‑level and system‑level strategies for proactive fault detection, data analysis, and alerting.
Migu Interactive Entertainment’s interview reveals how its cloud gaming platform leveraged AI, 5G, and standardized observability practices to pass both international and domestic AIOps assessments, highlighting the strategic importance of intelligent operations for business continuity in complex, distributed systems.
The paper proposes a business‑observability framework for a content‑production pipeline—illustrated by Bilibili’s workflow—by modeling archives as entities, assigning global AIDs for end‑to‑end tracing, and leveraging a Kafka‑Flink‑ClickHouse event‑streaming platform to monitor real‑time latency, bottlenecks, and safety audits across the entire production line.
The presentation outlines Alibaba Cloud's ACK container service observability framework, covering its architecture, key capabilities such as eBPF‑based tracing, GPU profiling, network diagnostics, storage monitoring, and FinOps integration, and demonstrates how these features support AI workloads, large‑scale production stability, and automated incident response.
Observability turns a black‑box application into a system by gathering logs, metrics, and traces, using alerts to spot anomalies, then linking trace IDs to logs; OpenTelemetry standardizes this with instrumented client agents, a Collector (receivers, processors, exporters), and backend storage, while Java agents, span propagation, exemplars, eBPF, and bundles like SigNoz or OpenObserve let teams choose between a custom OTel stack or a solution.
By deploying eBPF agents as DaemonSets that hook kernel network and performance events, the Xiaohongshu observability team extended cloud‑native monitoring from the application to the kernel, delivering real‑time traffic analysis and low‑overhead continuous profiling for C++ services, aggregating data into centralized collectors for dashboards, flame‑graphs, and rapid root‑cause diagnosis.
This article explains the concept of observability, contrasts it with traditional monitoring, outlines its benefits for system stability, reliability and performance, and provides practical guidance on building a full‑stack observability platform using logs, metrics, tracing and modern cloud‑native tools.
This extensive article presents a systematic approach to front‑end stability, covering observability systems, full‑chain monitoring, high‑availability design, performance management, risk governance, process mechanisms, and engineering practices to ensure reliable user experiences and business continuity.
The article explains how AI‑powered anomaly detection—using statistical 3‑sigma/Z-score methods, unsupervised machine‑learning like Isolation Forest, and deep‑learning models such as LSTM, Transformer and Pyraformer—overcomes the limits of threshold‑based monitoring by preprocessing data, reducing false alerts, and delivering high‑precision observability metrics.