AsiaInfo Won the National Runner-up in the 2020 AIOps Challenge

2020-08-17 AsiaInfo

On August 15, 2020, the final of the 3rd AIOps Challenge hosted by the China Computer Federation and undertaken by Tsinghua University, China Mobile and Sun Yat-sen University was held in Hangzhou. The intelligent operation team of AsiaInfo Telecom AI Labs overcame all the difficulties in the way and won the national runner-up.


1.jpg


The theme of this year’s AIOps Challenge was “Failure Discovery and Root Cause Location of Microservice Application System”, in which the data came from real microservice applications in a private cloud environment. 141 companies and college teams from all walks of industry, academia, and research fields participated the competition. After the preliminaries, two stages of semifinals and performance reviews, 6 teams were selected to enter the finals. The competition was quite fierce.


From the data of five rounds of preliminaries and two rounds of the semi-finals, the team composed of the AsiaInfo Telecom AI Labs has an average failure recall ratio of 85%, a precision ratio of 77%, and an average location time of 90s, ranking highly among all the 141 company and college teams.


The outstanding performance of AsiaInfo in this national competition was highly recognized by the customers of organizer China Mobile. A customer of a communications operator who was watching the competition at site said: “AsiaInfo has demonstrated an industry-leading level in the accuracy and versatility of AI algorithm. In the future, we hope to deepen cooperation and carry out joint innovation with AsiaInfo in the fields of big data, AI, and intelligent operations. ”


The relevant person in charge from AsiaInfo said that this award is not only a recognition of the AI technology and product research and development capabilities of AsiaInfo, but also will play a positive role in further promoting the commercial application of 5G data system products, AI and intelligent operations products in telecommunications and vertical industries.


Based on the actual business scenario of the microservice application system, AsiaInfo has developed a complete set of algorithm system based on the analysis of data characteristics, that is, from the perspective of the entire link, locate the root cause node of the calling chain through comprehensive analysis of the running status and calling relationship of each service, and further deduce and locate the actual failure root causes in an intelligent manner combined with platform index operating data and topological relationships.


At present, the microservice failure location capabilities have been productized and packaged, and some of the capabilities have been commercially applied in multiple operator customers of China Telecom, China Mobile, and China Unicom, helping the operations departments to locate the root cause of the failure in a timely and accurate manner.


2.jpg


AsiaInfo’s telecom AI innovation laboratory (Telecom AI Labs) was established in 2019. It is working on R&D, innovation and industrialization in the cross-field of communications and artificial intelligence, and it continues to improve AsiaInfo’s industry influence in the field of data science and telecom artificial intelligence through in-depth exchanges and cooperation with domestic and foreign industry, academia and research institutions.


AsiaInfo’s global intelligent operations platform (AISWare AIOps) is a global intelligent operations model and stipulation service and operations platform. The product is based on the AI platform algorithm model and focusing on the three directions of quality assurance, cost management and efficiency improvement. Oriented at the intelligent solution capabilities of operations scenarios such as failure discovery, diagnosis, disposal, prevention, decision support and resource optimization, the product will conduct componentized encapsulation to build reusable and evolvable operations model and stipulation and provide capabilities to the external world in the form of Open API, in order to support the rapid implementation of intelligent operations requirements, simplify the connection process, and reduce the cost of intelligence integration and addition of the operations system. This product has been widely used in the intelligent operations practices of many telecom operators, supporting nearly 40 AIOps scenarios, with an average daily intelligent service call of over 10 million.


AsiaInfo’s global artificial intelligence platform (AISWare AI²) is an infrastructure used to build large-scale intelligent services. As the hub of intelligence middle-ground of enterprises, it can provide step-by-step construction of algorithm models and full life cycle management services for the intelligence transformation of enterprises, and can help enterprises continuously accumulate their businesses into algorithm models to achieve the purpose of reuse, combination innovation, and large-scale construction of intelligent services. The product has been widely used in industries such as telecommunications, finance, transportation and energy, among which there are more than 30 provincial customers. Thousands of AI models of typical customers have been accumulated through the platform, with an average monthly call of intelligent service of nearly 100 million times.