Welcome toTencent Cloud + community, get more Tencent mass technology practice dry goods oh ~

This article was published in cloud + Community column by Tencent Cloud database TencentDB

Mobike is the world’s largest intelligent bike-sharing operation platform and mobile Internet of things platform. From its establishment to the world’s largest in less than three years, it generates more than 30TB of riding data every day and receives more than 30 million daily orders. What kind of database system is behind the massive data? It dominates over 180 cities in 9 countries including China, Singapore, The UK, Italy, Japan, Thailand, Malaysia, the US and South Korea, and operates over 7 million Mobikes. It has the most comprehensive cycling big data in the world. It provides intelligent mobility services to more than 200 million users worldwide.

From January 2015 to 2016, the entire technical team of Mobike was less than 50 people, and the backstage development team was only a dozen people. A dozen people face the rapid growth of the business, to solve so many problems, how to do? Mobike has wisely chosen Tencent Cloud.

Mobike started to connect to Tencent cloud’s database service at the end of 2016. The initial idea was straightforward and the company needed to save costs and focus on efficiency. I remember that Mobike calculated a very economical account at that time. It might be safe to build MySQL database by itself, but the development and maintenance of a series of data security, backup, monitoring and other functions required about 3+DBA. It would take more than half a year. Tencent cloud database with strong technical strength and massive services is the only choice for Mobike’s business.

TencentDB for MySQL database

  • International multiple ISO authentication, domestic information security protection level 3, network layer DDOS protection, SQL injection interception, data transmission service IPsec encryption connection
  • Covering connection access, load, storage engine and more than 70 important indicators of comprehensive monitoring
  • Disk array storage hardware high availability, real-time dual-system hot backup second-level switchover, two – and three-center remote DISASTER recovery
  • Full database o&M solutions such as backup, file back, monitoring, rapid expansion, and data migration simplify o&M work and focus on business development

Today, developers want their applications to be well built and effectively scalable, and to do so, they need to be able to use multiple databases and data models in the same application. This is proved in the product business of Mobike. Mobike selects the database according to the requirements of different businesses and has the Tencent Cloud database with a full range of products from relational to NoSQL to comprehensively guarantee the business scenarios of Mobike. : Tencent cloud MySQL is applied in bike order business, and Tencent cloud Redis is applied in data cache. MongoDB has always been the preferred database of major LBS service providers due to its friendly support for LBS query. Therefore, Tencent cloud MongoDB has become the only choice for Mobike users to search nearby bikes. At the same time, Tencent Cloud PostgreSQL is also used to cut the geographic fence map of parking positions.

Just like buying things, any product selected by users will be able to withstand comparison. Tencent cloud database products can win the trust of more customers like Mobike due to its high availability performance and 7X24 professional service. When Mobike used Tencent cloud MongoDB in the early stage, the team found that when the geoNear function was used to search for nearby objects, the problem of slow query often occurred. The pressure in the morning peak was 10-20 times that in the off-peak period, and the slow query was serious in the case of uneven coordinates, which was on the verge of avalanches. After analysis, the team found that these slow queries scanned too many point sets, and then passed the geoNear command into NearStage by optimizing the search volume for each turn and adding two parameters. This optimization is essentially a sacrifice of quality to get results as quickly as possible, but for most of the LBS services in China, complete rigor is not necessary these days. Then the optimization to help the thanks to improve the performance of the business more than 10 times, has a favorable review mo worship (detailed optimization method, the reference: cloud.tencent.com/developer/a…). .

In the high-speed development of the Internet industry, no matter the start-up or growth or mature Internet enterprises must keep their own business tenacious competitiveness, in order to be conducive to invincibility. As can be seen from mobike’s application of Tencent cloud database products and services, the continuous optimization and improvement of Tencent cloud database products and services will always be in line with the purpose of helping every customer find and solve problems. No matter the numerous basic database products well-known to all, Or the future of more SaaS products and will soon be C from the research of new database products CynosDB, Tencent cloud database customer expectations, has been working hard!

reading

Machine learning in action! Quick introduction to online advertising business and CTR knowledge

This article has been authorized by the author to Tencent Cloud + community, more original text pleaseClick on the

Search concern public number “cloud plus community”, the first time to obtain technical dry goods, after concern reply 1024 send you a technical course gift package!

Massive technical practice experience, all in the cloud plus community!