What Is a Cloud Time Series Database?

Chait Diwadkar
Chait Diwadkar

Widespread growth in digital technologies and interconnected tools has created a new world where a staggering amount of data can be collected across diverse sources. This includes many examples like stock prices, weather data, environmental readings, or data streams from sensors and Internet of Things (IoT) devices connected to all manners of equipment. As data collecting applications have proliferated, the sheer volume of time-series data has soared, creating the need to store it efficiently. The cloud time series database were developed to meet this demand. 

But what exactly is a cloud time series database and what makes it ideal for handling this type of data? How does a cloud time series database compare to other databases?

Answering these questions first requires a grasp of the underlying characteristics of time-series data, including how organizations can use it to stay competitive.

Understanding Time Series Data

Time-series data refers to datasets that are indexed according to time. Data points are captured over constant or evenly spaced time intervals, such as every minute or every millisecond. This kind of data is extremely useful in that it provides a record of how values or conditions change over time, allowing enterprises to identify trends and patterns and to use that information to predict future scenarios.

If a manufacturing facility is monitoring its systems and equipment to track performance, it is collecting time series data. The same is true for connected cars, which capture time-series data using multiple sensors and communication technologies. Applying analytics, these kinds of data streams can deliver powerful insights. With connected cars, vehicle performance and safety can be dramatically improved. Companies of all sorts can benefit by gaining a deeper understanding into their business operations and activities—helping to identify growth opportunities, drive efficiencies, enhance sustainability efforts, inform key decisions, and much more.

Today, the number of devices and sensors deployed are steadily increasing, and with vast quantities of streaming measurements continuously captured over really short time periods, timestamped data points can rapidly accumulate. These data sets are both extremely large and highly granular. All combined, these factors make time-series data challenging to ingest, store, and manage using traditional methods.

Key Advantages of Cloud Time Series Databases

Cloud time series databases evolved out of the need for a solution that addresses the unique requirements and constraints of time series data. As IoT data volumes have skyrocketed, driving data storage demands higher, cloud time series databases provide an architecture designed for write throughput performance—optimized for real-time ingestion of high-velocity, high-volume time-series data inputs.

With more enterprises exploring how to use their data to deliver actionable insights, cloud time series databases offer specialized functions tailored to the properties of timestamped data, all while drawing from the cloud to deliver more flexibility. To extract value (i.e., insights) from time series data, cloud time-series databases provide efficient storage and retrieval, including rapid query response times and enhanced data access.

Another important feature: cloud time-series databases are designed to scale as data storage and workload demands continue to rise. And they employ organizational approaches for better managing large-scale datasets. This includes compression, which reduces data size, as well as downsampling—a system task that minimizes the granularity of data by aggregating multiple data points into a single value. 

Some of the ways that cloud time-series databases are distinct from other kinds of databases include:

  • Cloud time-series databases are purpose-built to ingest, organize, store, and retrieve data based on timestamps.
  • Cloud time-series databases are configured to support operations according to a time-centric framework.
  • Cloud time-series databases give users the flexibility to define their data retention polices with respect to archiving or destroying older data.

Compared to relational databases, which store diverse data types in predefined, structured tables and are designed to recognize data relationships within that arrangement, cloud time-series databases store data in a compressed format with retrieval, queries, and processing optimized for time-ordered data. NoSQL databases have a flexible data model and the ability to store a range of data types in varying formats. However, NoSQL databases may be less equipped to handle the requirements of time-series workloads efficiently. Generally, relational and NoSQL databases are not built for the ingestion, query, and storage demands of massive time-series data sets.

Use Cases

From industrial plants and water delivery systems to financial trading, logistics, smart buildings and more, a growing number of applications are using cloud time-series databases to leverage the possibilities of time-series data for improved decision-making.

For example, an industrial plant that collects time-series data from assets and production machinery using IoT tools can benefit from the analytical capabilities of a cloud time-series database to optimize the performance of equipment, such as a steam boiler. Insights from a cloud time-series database can help that plant to detect anomalies in its processes, anticipate potential issues, and facilitate preventative maintenance. Overall, these actions increase asset life, minimize unexpected downtime, and improve plant productivity.

Similarly, a cloud time-series database can be invaluable to water utilities for analyzing data collected across water distribution systems to detect water losses, pinpoint leaks, and prioritize actions to take, like which pipes should be repaired or replaced. As with so many other examples, cloud time-series databases are integral to companies for deriving value from the data they collect and using it to their advantage.

  • Chait Diwadkar
    Chait Diwadkar

    Chait Diwadkar is Director of Solution Engineering at TDengine. Prior to joining TDengine he was in the biotechnology industry in technical marketing, professional services, and product management roles and supported customers in pharma, medical devices and diagnostics on analytical chemistry, and genetic platforms.