产品系统设计Todo Products Todolist

The system design in production

1. from zero to one
   1. sgpass.info
   2. aicodereviewer
2. projects
   1. slora
   2. distributed-systems
3. order fee

   1. Scenario

      1. Query Order Optimization
         1. DB Index Degsing, escpecially for time
         2. start_time > x and end_time <x how to make an efficient index
            1. 假设我有一个含 order_id，start_time和end_time的查询 where order_id = x and start_time > 100 and end_time < 200，mysql的index该怎么优化呢？=> CREATE INDEX idx_order_time ON your_table_name (order_id, start_time, end_time);
         3. how to query by order_id and order_sn while there is sharding?
            1. currently they do this by TIDB query with index
         4. 每年都有几billion的order，如何去分表分库，并且如果按order_id来做为sharding key后，如何在没有制定order_id的情况下去做索引，当下业界是如何实现的？
            1. 分表分库是处理大规模数据的有效方案。通过使用全局索引、广播查询或中间件等方法，可以在没有指定 order_id 的情况下进行高效查询。
            2. 看起来目前是根据data studio
4. settlement
5. seller wallet
   1. status monitors system