We media: Yizhuan
PolarDB-X is designed to solve database bottlenecks, such as challenges with mass data storage, ultra-high concurrent throughput, large table performance, and complex computations. Its performance has been proven in production environments by Tmall during every Double 11 Global Shopping Festival and by Alibaba Cloud customers across various industries. PolarDB-X helps enterprises accelerate their digital transformation. This topic describes how Yizhuan uses PolarDB-X to overcome its business challenges.
- Industry: We media
- Website: Yizhuan
About the customer
Changsha Yingzhi Information Technology Co., Ltd. is a professional provider of big data services for new media. Its well-known brand, Yizhuan, utilizes new media data mining technology and Natural Language Processing (NLP) algorithms. Yizhuan provides content creators and advertisers with comprehensive and precise big data analytics services and big data architecture solutions.
For many years, the company has focused on innovation. It has developed multiple tools for new media operations. These tools help new media professionals work more efficiently, break down information silos, and continuously optimize the content creation ecosystem. The company collaborates with others in the industry to promote mutual growth.
Business challenges
- High concurrent write pressure on the data ingestion service, reaching approximately 5,000 Transactions Per Second (TPS).
- Complex business queries that span a wide time range.
- Single tables with billions of rows, requiring multi-dimensional queries across hundreds of millions of rows.
- Rapidly growing data volume, requiring terabytes of storage.
- A requirement for data updates and strong consistency for transactions.
Solution
- A distributed database architecture that combines PolarDB-X and RDS is used to handle high-concurrency writes from the data ingestion service.
- PolarDB-X and PolarDB are used for business scenarios that require queries over a wide time range.
- The mass storage capability of PolarDB is used to meet data storage requirements of 5 TB to 10 TB.
- PolarDB supports complex queries over wide data ranges while also supporting transactions.
Customer value
- The data ingestion service required high-concurrency database writes, which a standalone MySQL database could not support. The combination of PolarDB-X and RDS met these high-concurrency write demands and increased the overall business load capacity.
- The ability to upgrade and add compute nodes in minutes resolves resource bottlenecks during peak business hours.
- Highly elastic mass storage solves the storage bottleneck for large amounts of data.
- The I/O and SQL query optimizations in PolarDB address the pain point of multi-dimensional queries over a wide time range.
Testimonial
"Our business requires high-concurrency writes. The distributed database solution of PolarDB-X and RDS perfectly meets this need. We also run many complex, multi-dimensional queries over wide data ranges. The high-performance read capability of PolarDB is the ideal solution for this scenario. By combining the high-concurrency writes of PolarDB-X with the high-performance reads of PolarDB, we have completely solved our pain points with data ingestion and wide-range queries."
— Liu Tao, Technical Director, Changsha Yingzhi