High-performance Architecture---storage performance optimization for websites

Large amounts of data read and write to the disk access pressure , many times the disk is the system's most serious bottleneck .

• Mechanical hard drives vs solid-state drives

The mechanical hard disk drives the head arm through the motor, drives the head to the specified disk location to access the data, since each access to the data needs to move the head, so the mechanical hard disk in continuous data access and random access performance gap is huge.

SSDs, also known as SSD or flash hard drives, have fast data access and low power and noise.

• B + trees vs lsm trees

In order to improve the data access characteristics, the file system or database system usually stores the data after sorting, speeding up the data retrieval speed. The traditional relational database approach is to use B + numbers.

Currently, the database uses a B + tree with a level two index. Today, many NoSQL products use LSM trees as their primary data structure.

When a read operation is required, the search is always started from the in-memory sort tree, and if it is not found, it is searched from the sorting tree order on disk.

A data update on the LSM tree does not require disk access and can be done in memory, much faster than a B + tree. When data access is written primarily, while read operations focus on recently written data, using the LSM tree can greatly reduce the number of disk accesses and speed up access.

• RAID vs HDFS

RAID (Redundant array of Inexpensive disks) technology is primarily designed to improve disk access latency and enhance disk availability and fault tolerance. By using RAID technology, it realizes concurrent read and write and data backup of data on multiple disks.

High-performance Architecture---storage performance optimization for websites