Time Series Database Tutorial

The modern world is filled with situations where the timing sequence is critical to effective analysis. The Internet of Things (IoT) is generating high volumes, continuous flow of metrics obtained through measurement of devices. Stock markets are generating a high volume of trades on a second by second basis. Computer systems are generating high volumes of data about there status-bar such as jobs processed, memory used and computing cycles expended. High volume streams of data need to be rapidly captured, stored and analyzed. This is a big challenge.

What is a Time Series Database?

A Time Series DataBase (TSDB) is a DataBase Management System (DBMS) optimized to store and access high volumes of data which contains timestamps, an attribute which specifies date and time to a detailed level. Time Series DataBases provide distinctive capabilites:

• Scale: database can quickly store huge volumes of time series data.
• Efficient Storage: data is stored efficiently - in compressed format specifically tailored to series of measurements.
• Analytical Functions: enables aggregation, odering, trending, and pattern finding in time series data.

Who Provides Time Series Databases?

Multiple vendors provide Graph DBMS Software. DB-Engines publishes a ranked list of graph DBMS. InfluxDB is the market leader in TSDB according to DB-Engines research. The top five in that list as of June 9, 2020 are:

• InfluxDB : score 21.18
• Kdb+ : score 5.86
• Prometheus : score 4.59
• Graphite : score 3.63
• RRDtool : score 2.90

In addition to the above suppliers, consider DBMS that have been improved to support time series such as: Teradata, Redis and Informix.

Time Series Data References and Links

Check out these Time Series Data links:

• An Introduction to Time Series Databases : Kryzstof Ksiazek, 2019, Several Nines
• When Should I Use a Time Series Database? : Kathy Farmer, 2018, The New Stack
• Key Concepts and Features of Time Series Databases : Zhaofeng Zhou (Muluo), 2019, Alibaba Cloud