Vertical, Hortizonal sharding, read/write separation , NewSQL....
Created#More Posted time:Jan 30, 2019 17:08 PM
Vertical, Hortizonal sharding, read/write separation , NewSQL.... There is no one size fits for all
....Most relational database products use B+ tree indexes. When the data volume exceeds the threshold, the increase in the index depth leads to an increased disk I/O count, the substantially degrading query performance.......
Although table sharding can resolve the performance problems caused by massive data, it cannot resolve the problem of slow responsiveness caused by excessive requests to the same database. For this reason, database sharding is often implemented in horizontal sharding to handle the huge data volume and heavy access traffic. Read/write separation is another way to distribute traffic. However, you must consider the latency between data reading and data writing when designing the architecture.
NewSQL with a brand new architecture resolves this problem in a way that is different from that of the sharding middleware:
Many may ask what is the difference between HTAP vs NewSQL , though :)
Sharding consists of the following processes: statement parsing, statement routing, statement modification, statement execution, and result aggregation.
Community -> Blog -> Breaking the Limits of Relational Databases: An Analysis of Cloud-Native Database Middleware (1)