Help customers' digital transformation and build a new operation and maintenance system

author ： Jinbiao

      Driven by the new generation of digital technology , The enterprise moves from manufacturing to “ Zhi Zhi ”,IT The pressure of operation and maintenance has doubled , It has brought great challenges to operation and maintenance services and platforms , Hua Wei Yun “ Deterministic operation and maintenance ” The scheme helps enterprises find new breakthroughs in the operation and maintenance process , Build a new operation and maintenance system , Provide security for customers on the cloud 、 major 、 Efficient service .
      Huawei cloud has become an important engine for customers' digital transformation
      Huawei cloud customer is a large state-owned communication infrastructure service enterprise , While carrying out traditional communication service business , The company is actively carrying out scientific and technological innovation .
      Hua Wei Yun SRE When the engineer talks with the customer's project leader , Combined with the business characteristics of the enterprise , Let business capabilities be born in the cloud , Longer than clouds . Born in the cloud is an intelligent operation and maintenance platform based on Huawei cloud to build enterprise operation and maintenance , Being good at cloud is to make full use of the advantages of cloud to help customers' applications and business development , Bring customers' digital construction and business intelligence upgrading to a new stage .
      After the formal launch of the project support , Hua Wei Yun SRE、CRE、TAM、RDPM Members move in quickly , Cooperate with customers to systematically analyze the business , Identify full risk 、 Rely on monitoring best practices to assist customers in rectification 、 Output reliability reinforcement scheme and disaster recovery scheme , Promote quasi production / The production environment is complete MAS double AZ The dual active disaster recovery scheme has been implemented .
      The Huawei cloud team cooperates with the customer to evaluate the business system from the instance status 、 edition 、 Capacity performance 、 Overload protection 、 Operational compliance 、 Business module reliability 、 Failure explosion radius 、 Analyze and evaluate the escape and emergency plan , Explicit risk items , Benchmarking the risk elimination plan with customers and implementing . For risks that cannot be eliminated , Regularly carry out the failure drill of the emergency plan , Ensure that the fault can be quickly recovered .
      “ Deterministic operation and maintenance ” System enabling digital operation and maintenance transformation
      In terms of business inspection and change control , Focus on the hidden trouble of customer hardware 、 Background instance status check 、 In depth inspection of cloud services , Identify risks and rectify them , For various changes , Intra domain review + Review of major changes shall be double checked .
      In the full link monitoring system , According to the customer's business characteristics , Take stability as the core demand , Perform multi round threshold and alarm configuration optimization , Continuously optimize the monitoring indicator threshold according to the business operation .

      In the system capacity analysis , The current business system capacity can cope with single AZ Business continuity of level-1 resource failure scenarios . The hierarchical level analyzes the related information according to the business modules Redis、MongoDB、RabbitMQ、Kafka、CCE、ELB、NAT、ECS,8 Capacity assessment of key cloud services for core businesses .
      In the disaster recovery drill , Set the drill scenario , Drill the fault handling process respectively 、 Surprise random failure drill 、 Single component failure drill 、 Whole system fault recovery drill, etc , And through practice , Be familiar with fault emergency plan , Shorten the recovery time of existing network fault services .
      On the structural reinforcement , Customer architecture reinforcement is achieved through three stages . In the first phase , Reliability reinforcement of key examples 、 many AZ Deploy ： Active and standby instances span AZ Transform and implement redundant clusters Redis、MongoDB Span AZ Disaster recovery deployment , Implement multi cluster through content DNS Perform active / standby switching ; In the second phase , Emergency applet disaster recovery environment disaster recovery ： Span Region Deploy the wechat applet disaster recovery environment , Realize fast disaster recovery switching of core business . The Huawei cloud support team cooperates with customers to implement the scheme and complete joint exercises , Ensure the smooth launch of the emergency applet ; In the third phase , Based on Huawei cloud MAS Multi active and highly available solutions （ Consumer cloud multi live disaster tolerance incubation scheme ）, Output the main business across AZ Double active disaster recovery scheme .
      After a series of efforts , Ultimately help customers achieve IT Unified operation and maintenance management 、 Unified protection 、 Unified monitoring , Establish comprehensive safety monitoring 、 Notification, early warning and emergency disposal mechanism , Greatly improved the quality and efficiency of operation and maintenance .

      Huawei cloud continues to iterate and restructure through the industry operation and maintenance indicator system , Realize intelligent operation and maintenance to protect the enterprise's digital transformation business . Through a complete set of operation and maintenance guarantee solutions , It ensures the smooth operation of customers' massive devices and business applications , Huawei cloud continues to be guided by business support , Guided by the operation and maintenance system , Supported by the operating platform , Continuously improve customer operation and maintenance management and service level , Ensure sustainable business development , Fully meet customer needs , Refine service levels , Help customers succeed in digital transformation .