Alibaba's Cheng Chao: The ultimate development of the monitoring system is to achieve intelligence

[51CTO.com original article] Seven years of hard work, a gorgeous transformation. Since 2012, 15 summits have been held for 6 consecutive years, gathering a large number of technical experts, and have learned from a wide range of knowledge and accumulated experience. The 2018 WOT Global Software and Operation Technology Summit has set sail, focusing on 12 core hotspots, bringing together 60 first-line experts from home and abroad to create a high-end technology feast! At the 2018 WOT Global Software and Operation Technology Summit on May 18-19, Mr. Cheng Chao, a senior technical expert at Alibaba, will give a wonderful speech entitled "Building a Super-Large Second-Level Monitoring Platform" to the majority of technical colleagues.

Mr. Cheng Chao has nearly ten years of experience in operation and maintenance system development. He currently works in Alibaba Infrastructure Business Group and is responsible for the construction of Alibaba Group's monitoring platform. He led the construction of the first generation of Alibaba CMDB system. In recent years, he has focused on the monitoring field and is now responsible for the monitoring platform that covers all Alibaba business groups. He has experience in building large-scale, distributed monitoring systems.

[[226871]]
Cheng Chao, Senior Technical Expert at Alibaba

The evolution of Alibaba's three generations of monitoring systems

During the Double 11 Global Shopping Festival last year, Tmall's total transaction volume for the day was 168.2 billion yuan, with a peak transaction volume of 325,000 per second. Alibaba's third-generation monitoring system was used to monitor these second-level data. Looking back on the evolution of the data monitoring platform, Mr. Cheng Chao gave a general description: the first-generation monitoring system was mainly built on open source systems, and soon encountered problems with stability and scale; the second-generation monitoring system solved the problems of stability and large-scale application, but the scope was mainly focused on monitoring at the system and application levels, lacking monitoring of key points such as links and businesses; the third-generation monitoring system is mainly for dev Ops monitoring systems, focusing on new business requirements such as intelligence, standardization, and full business links.

Stability is the most important indicator for monitoring systems

Mr. Cheng Chao admitted that the most important foundation of the monitoring system is its own stability. The team improves through continuous practice and trial and error, forming a virtuous cycle. To ensure the stability of the monitoring system, it is necessary to establish a feedback mechanism of self-monitoring + normalized exercises. The problem of self-monitoring is relatively simple to solve. In the early days, Alibaba used an open source monitoring system to monitor its own monitoring system, and later gradually evolved into a system that monitors itself. The advantage of this is that you are the user and can continuously optimize the product experience during use.

The so-called exercise routine is Alibaba's own unique monitoring mechanism, which is usually carried out on a monthly basis to regularly exercise the performance of each module in the event of a failure (such as a computer room network outage, a sudden increase in log volume, and abnormal CMDB data synchronization). In addition, real-time regression must be run online to continuously verify data to ensure that there are no logical errors after each change.

Three Development Trends of Monitoring Systems

As an expert in the field of monitoring platform development, Mr. Cheng Chao has a deep understanding of the development trend of monitoring platforms based on his work field. He believes that in the large-scale scenario of Alibaba Group, the development of monitoring systems will have several directions: the first is standardization, which requires pre-definition of standardized indicators and templates. The goal of achieving standardization is to lower the threshold of basic monitoring and provide a data foundation for intelligence. The second is integration, including horizontal and vertical connection of monitoring data, organizing data of all dimensions with business as the core, and the goal of integration is to show the full picture of the business and shorten the time of fault location. The last is intelligence, which has the most possibilities to explore, and its goal is to make the monitoring system the brain of operation and maintenance, and ultimately achieve unattended operation.

Want to know more? On May 18-19, the world's most noteworthy IT technology feast will be held at the Beijing JW Marriott Hotel. The 2018 WOT Global Software and Operation Technology Summit is an important platform for you to discover new ideas, explore the latest ideas, and expand your network.

At present, all our tickets are on sale. Please note that the earlier you buy your tickets, the bigger the discount! Communicate with KOLs up close and present a different "hero feast"!

Click on the official website for details: wot..com

[51CTO original article, please indicate the original author and source as 51CTO.com when reprinting on partner sites]