High Availability Series Part 4 - Summary

Previous link

1. Part 1 of a highly available series - Overview - Dongfeng Weiming Technology Blog (e-whisper.com)
2. High Availability Series Article 2 - Traditional Layered Architecture Technology Solution - Dongfeng Weiming Technology Blog (e-whisper.com)
3. Article Series No. 3 - NGINX High Availability Implementation - Dongfeng Weiming Technology Blog (e-whisper.com)

VI Summary

High Availability HA (High Availability) is one of the factors that must be considered in the design of the system architecture, which usually refers to reducing the time that the system cannot provide services through design.

Methodologically, high availability is throughredundancy+Failoverto achieve.

The recommended high-availability solution of the entire manufacturing system architecture passes through each layerredundancy+FailoverTo synthesize the implementation, specifically:

1. Client layer arrive Load balancing layerHigh availability is achieved through redundancy of the load balancing layer, and the specific practice is keepalived + NGINX automatic failover
2. Load balancing layerarrive App service layerHigh availability is achieved through redundancy of the application service layer, and the common practice is the survivability detection and automatic failover between nginx and the application server
3. App service layerThe high availability to the database “read” is achieved through the redundancy of the read library, the common practice is to perform “master-slave replication, read-write separation” through the database layer, and the database connection pool of the application service layer configures the IPs of multiple read libraries to ensure automatic failover
4. App service layerThe high availability of database “write” is achieved through redundancy of write libraries, and the specific practice varies according to the type of database.

appendix

Appendix 1: Comparison table in English and Chinese

Chinese	English	English abbreviation	Notes
Availability	Availability
Downtime	Downtime
Mean Time Between Failures (Or Mean Time Between Failures)	Meantime between failure	MTBF
Average time to repair	Mean time to repair	MTTR
Failover	fail-over
Load balancing	load balance
Active-slave switchover (active to standby)	Active-passive
Duplex switching	Active-active
Redundancy	redundant
Browser	browser
Client	client
Single application	monolithic application
Web server	Web Server
Polling	Round Robin	RR
Weighted Cycle	Weighted Round Robin	WRR