Preface many bloggers asked me to modify the title. Article It does not have much to do with large websites and architectures, so I decided to modify the title. As mentioned above, we talk about the stretch tree. This is about the isam tree. The isam tree was proposed by IBM in 1964.
Index sequential access method (Isam, indexed sequential access method) is initially IBM Developed by the company File System , You can record any access consecutively (in the order they enter) or randomly (based on the index. If Code Interested, you can refer to the mysql-3.23.35 source code, the code inside has reference value. -- I will not explain the code. I will make a series of articles for isam if I have time. Isam Structure
Equivalent to multi-cross balancing tree
Tree height (reducing hard disk I/O times)
Multiple node records (read more data at a time)
Isam search
Isam search efficiency
Isam Index insert data
Isam used the only engine (isam engine) before FoxPro/dbasems accessmysql 3.23 -- later replaced by MyISAM (not isam tree), Berkeley dB advantages: simple, easy to implement problems: overflow page imbalance, so we need to rebuild index?
To sum up, I did not explain the specific code layer. I found that there are very few isam materials. So I will download it. Source code Look, if you want to have a deeper understanding, I suggest you can download the mysql-3.23.35 source code for a good study, the early version of the code is still relatively small. Here is a reference C-isam manual.
Recommendation