Intelligent Operations Sessions at the 2018 Hangzhou Yunqi Conference

In September 2018, the Hangzhou Yunqi Conference was held in Hangzhou Yunqi Town with the theme “Driving Digital China,” offering over 170 cutting‑edge sessions, including a highly attended 200‑person “Intelligent Operations” track.

Alibaba Intelligent Operations System Construction – Liu Guohua, Alibaba Group researcher, described Alibaba’s evolution from manual to platform‑based and finally intelligent operations, emphasizing the need for machine intelligence to meet high‑complexity, high‑security, high‑reliability, and high‑efficiency demands of modern infrastructure.

Liu outlined four guiding mindsets – system thinking, baseline thinking, security thinking, and global thinking – and explained how Alibaba integrates machine learning, optimization algorithms, and domain expertise to achieve automated, intelligent maintenance across supply chain, server, cluster, and application operations.

Robust Design of a Reverse Supply Chain Network Planning – Prof. Zhang Zhihai, Tsinghua University, presented a photovoltaic battery recycling network case, using robust optimization to model uncertainties (price, demand) and a step‑wise algorithm to ensure cost‑effective, resilient designs despite parameter variations.

Intelligent Application Operations – Huang Xinyi, Alibaba senior technical expert, introduced Alibaba’s largest operations platform transformation, DevOps best practices, and an “unattended release” solution that leverages machine learning and optimization to detect and block anomalous changes before they cause failures.

MSP Leads Intelligent Operations Transformation – Li Yun (Brad Lee), CEO of Beijing Bespin Cloud Technology, discussed how third‑party Managed Service Providers (MSPs) are driving a new era of AI‑Ops by aggregating massive operational data, experience, and algorithms to build intelligent, data‑driven services.

Data‑Driven Operations Building Intelligent Capability – Sun Yonghua, Alibaba operations expert, explained the development of Alibaba’s big‑data SRE framework, introducing the DataOps concept and three key application scenarios: knowledge graph (search & ChatOps), intelligent monitoring, and operational optimization.

Intelligent Large‑Scale Cluster Operations – Sug Xiaoxiang, senior Alibaba technologist, described how automated and intelligent methods improve stability and reduce costs for massive clusters, detailing planning‑phase risk control via gray‑scale models and anomaly handling through data‑driven decision making.

Intelligent Data Center Operation – Jiao Jing, Alibaba senior technical expert, outlined the evolution from data‑centric to automated to intelligent data‑center management, highlighting unified monitoring, AI‑based optimization, and the shift from reactive to proactive maintenance.

Intelligent Data Center Supply Brain – Zhu Wanyi, Alibaba senior technical expert, presented a supply‑chain brain that uses predictive analytics, operations research, and AI to optimize resource allocation, reduce costs, and mitigate supply risks across the data‑center lifecycle.

Intelligent Fault Management – Wang Zhaogang, Alibaba senior technical expert, shared Alibaba’s “Intelligent Baseline” framework that combines time‑series decomposition, machine learning, and online data‑warehouse queries to pinpoint anomalies and accelerate incident response.

Intelligent Emergency Collaboration – Guo Rui, Alibaba technical expert, described an IM‑based robot system that leverages big‑data analysis and intent training to coordinate emergency handling, ensuring rapid, unified response across complex ecosystems.

The conference concluded by emphasizing Alibaba’s ongoing “DC Brain” initiative, which integrates AI and domain expertise to create a self‑driving data‑center capable of autonomous delivery, proactive optimization, and continuous learning.