PingCAP’s Top 10 Posts of 2021

2021 has been a fruitful year for PingCAP. We released TiDB 5.0, which offers users a one-stop HTAP database solution, and we announced the public preview of TiDB Cloud. With TiDB Cloud, users can now easily use TiDB on Amazon Web Services and Google Cloud to quickly build modern, mission critical applications.

Our writers were busy this year, too. We published more than 90 technical articles and case studies focusing on advanced data technologies, industry pain points and solutions, and many other technical topics. 43 of them have been welcomed and published on prominent sites such as DZone, The New Stack, and Google Cloud, and several articles made the front page of Hacker News. 

Which articles were the most popular? Based on our own page views, here are the 10 articles that interested our readers the most.

1. How to Efficiently Choose the Right Database for Your Applications 
   
   It’s not easy to find the right database for your applications. This post shares some useful criteria and decision models to help you choose between different options. 
   
2. How We Trace a KV Database with Less than 5% Performance Impact 
   
   As a key-value database, TiKV has much higher performance requirements than a regular application, so tracing tools must have minimal performance impact. This post tells you how to trace TiKV requests while impacting performance less than 5%.
   
3. Linux Kernel vs. Memory Fragmentation (Part I) 
   
   Memory fragmentation is a long-standing Linux kernel programming issue. In this post, we introduce some common extensions to the buddy allocator that helps prevent memory fragmentation. 
   
4. tidb-lite: A Simpler Way to Unit Test Golang Database Code 
   
   When you unit test database-related code, go-sqlmock is a commonly used tool, but it has drawbacks. This post discusses its limitations, introduces tidb-lite, a simpler way to unit test such code, and shows examples of tidb-lite in action.
   
5. Why This MySQL Alternative Beats Vitess and CRDB in Scaling Out Our Databases on K8s 
   
   Ninja Van is a logistics giant in Southeast Asia. This case study shares Ninja Van’s pain points in their applications and describes why they finally chose TiDB over Vitess and CockroachDB. 
   
6. Reducing Real-Time Query Latency from 0.5 s to 0.01 s with a Scale-Out HTAP Database
   
   This is also a case study on Autohome, the leading online destination for automobile consumers in China. It shares how TiDB helps Autohome scale out their database and achieve real-time analytics. 
   
7. How to Efficiently Stress Test Pod Memory 
   
   This is a tutorial that helps you to troubleshoot issues you might encounter when you use StressChaos, a Chao Mesh tool that allows you to inject CPU and memory stress into your Pod. 
   
8. Linux Kernel vs. Memory Fragmentation (Part II) 
   
   This is a sequel to Linux Kernel vs. Memory Fragmentation (Part I) which ranks the third on today’s list. In this post, we explain the principle of memory compaction, how to view the fragmentation index, and how to quantify the latency overheads caused by memory compaction.
   
9. Five Principles that Guide TiDB and PingCAP (Part I) 
   
   As a product evolves, sometimes the features you leave out are more important than the ones you include. In this post, Max Liu, the CEO of PingCAP and TiDB’s principal product manager, shares his philosophy behind TiDB’s evolution, where TiDB has been, and how we finally got there. 
   
10. TiDB Operator Source Code Reading (V): Backup and Restore 
    
    When you maintain a database, backup and restore are two of the most important and frequently used operations. In this article, we talk about the core design logic of TiDB Operator’s backup and restore features and how to implement them.  

Edited by: Tom Dewan