Time series Database Selection--essentially column storage, B-tree index, or inverted index in search engine

The time series database is the largest and most widely used. This type of storage is commonly referred to when people talk about time series databases. Depending on the underlying technology, it can be divided into three categories.

• Straightforward file-based storage: RRD Tool,graphite Whisper. This type of tool is attached to the Monitoring alarm tool, and there is no regular database engine at the bottom. Simply there is a binary file structure.
• Build based on k/v database: Opentsdb (HBase-based), Blueflood,kairosdb (Cassandra based), Influxdb,prometheus (based on LEVELDB)
• Build on relational database: Mysql,postgresql can be used to save time series data

For a comparison of its underlying principles, it is best to read the following three articles!

Turn from:

Http://www.infoq.com/cn/articles/database-timestamp-01

http://www.infoq.com/cn/articles/database-timestamp-02

http://www.infoq.com/cn/articles/database-timestamp-03

Time series Database Selection--essentially column storage, B-tree index, or inverted index in search engine