AWS Time-Series Database: Understanding Your Options

Written by Carlo Mencarelli

Over the years, Amazon Web Services (AWS) has become a leading provider of cloud database hosting services. They offer everything from general relational databases to specialized ones like ledger and graph databases.

One type of specialized database that has been gaining significant popularity over the years is the time-series database (TSDB). DB-Engines, an initiative by Solid-IT to collect and present information about database systems, ranks time-series databases as the second highest change in popularity for the last several years.

With the growing importance of time-series data and options available, there are plenty of options available for users of AWS. However, choosing the right solution for your specific needs can be challenging. I hope to clear up some of the confusion and make the choice easier by the end of this article by:

• Providing a clear understanding of time-series databases

• Exploring the various time-series database options available on AWS

• Comparing these possibilities to help you decide what's right for your organization

What Is a Time-Series Database

As mentioned, time-series databases are specialized database systems built to handle time-stamped or time-series data efficiently. Some examples of this data would be server metrics, sensor data, stock market information, or weather data. For more information about time-series data, we have a whole blog you can read: Time-Series Data: What It Is, and How to Use It.

As you may have learned from our blog above, time-series data is time-centric and normally append-only. A TSDB leverages these facts to optimize storage and compute operations and can efficiently handle a large number of transactions in a very short time.

Queries on time-series databases typically involve date ranges, which means the query process needs to be optimized to handle these types of requests. TSDB systems include these optimizations as part of the engine. Some systems can even downsample or archive the data as it becomes older.

If you want to know more about TSDB systems, our recent blog, Time-Series Database: An Explainer, provides more detail.

What you need from a time-series database

When choosing a time-series database, there are several key properties to consider:

• Scalability: A good TSDB should be able to handle increasing volumes of data and query loads without significant performance degradation.

• Maintainability: The database should be easy to manage, update, and troubleshoot. It should include features like automated backups and archiving.

• Reliability: Many users of TSDBs are building new systems from scratch, so having the data when needed is crucial for success.

• Usability: The database should offer a powerful and intuitive query language that allows users to easily retrieve and analyze time-series data without adding a new language that the developers have to learn and become proficient with.

AWS Time-Series Database Options

Several options are available for time-series database solutions on AWS. These can be broadly categorized into AWS-native solutions and third-party options that can be deployed on AWS infrastructure. Let's explore some of the key options.

Amazon Timestream is AWS's purpose-built time-series database service. It comes in several flavors, depending on your needs, which we’ll cover below.

You can also use Amazon RDS or Amazon Aurora, although they were not built specifically for time-series data. RDS offers typical relational databases, while Aurora offers Amazon’s proprietary versions of MySQL and PostgreSQL with a serverless option available.

Timescale, built on AWS infrastructure, offers another option. It is a full time-series database solution based on PostgreSQL, so you get TSDB performance with the familiarity of PostgreSQL.

AWS solutions

AWS Timestream

Amazon Timestream is AWS's purpose-built time-series database service. It offers two variants: Timestream for LiveAnalytics and Timestream for InfluxDB.

Timestream for LiveAnalytics is Amazon’s proprietary TSDB. The database falls under their serverless offerings, which means it will automatically scale up and down based on performance requirements. It also offers the standard 99.99 percent uptime guarantee that many AWS services do. 

However, the service has some significant limitations, including limited SQL support, specifically the inability to do joins. Additionally, it is a proprietary solution, which means the service is only available on AWS and has no community development.

Timestream for InfluxDB is based on the open-source implementation of InfluxDB, currently compatible with the 2.7 InfluxDB service. The service is fully managed by AWS, meaning you don’t need to worry about installation and some administration tasks. Timestream for InfluxDB is also beholden to the same problems that running InfluxDB on your own is. We’ve written about some of what InfluxDB has gotten wrong over the years, but to name a few: two backend rewrites, two major API changes, and a multiplying product catalog of similarly named services causing confusion.

RDS

RDS is Amazon’s basic database service. It offers several different engines, such as MySQL, PostgreSQL, and Oracle. It does not offer specific TSDB features, but you can leverage open-source options such as the pg_partman PostgreSQL extension to assist with partitioning the time-series data.

This level of flexibility is nice but also adds management and development overhead. You can find more open-source TSDB options in an article we recently published.

Amazon Aurora

Aurora is similar to RDS in that it provides MySQL and PostgreSQL-compatible databases. The difference is that Aurora offers more cloud-specific features, such as storage scaling, point-in-time recovery, and more.

While Aurora does provide a better administration experience, like RDS, it doesn’t offer any TSDB-specific features or capabilities. The added cost and capabilities may not exactly add up, as we explored in an earlier article: Benchmarking Amazon Aurora vs. PostgreSQL.

Timescale on AWS

Timescale offers a few products in the TSDB space: an open-source TSDB PostgreSQL extension that is free to use (TimescaleDB) and a fully managed solution that offers advanced features and support (Timescale Cloud, with TimescaleDB at its core).

TimescaleDB is built on top of PostgreSQL, which provides a familiar interface and features of PostgreSQL but adds a powerful TSDB over it. In several tests, we’ve observed Timescale surpassing InfluxDB with time-series data. Timescale supports columnar compression for even more efficient data storage, ultimately saving time and money on storage and data transfer.

Conclusion

Time-series databases have become critical tools for teams dealing with large volumes of time-series data. As we've seen, they offer high-performance solutions specifically designed to handle the unique challenges of time series data, such as high write throughput, efficient storage, and optimized time-based queries. There are a lot of choices in the ecosystem for TSDBs, with the information above, you should be well-equipped to make the decision best for you and your organization.

You can host all types of PostgreSQL workloads in Timescale. But especially if you have time-series data, Timescale will suit you better than other AWS time-series databases. Plus, by choosing Timescale Cloud, you’ll enjoy the following benefits:

• Our mature cloud platform is supercharged PostgreSQL (simpler, faster, and more cost-effective). If you work with time-series data, you know how quickly your data will grow—avoid hitting that scalability wall, and definitely don’t be scared to pay for top performance every time you check your monthly bill.

• Enhanced performance with less compute and storage resources, thanks to specialized optimizations tailored for time series and analytics workloads. For a 1 TB dataset with almost one billion rows, Timescale outperforms Amazon RDS for PostgreSQL, with up to 44 percent higher ingest rates and queries running up to 350x faster.

• Substantial storage cost savings: our advanced compression algorithms enable a remarkable 90 percent reduction in disk storage. You can tier your older data to object storage built on S3 for an extra savings boost while remaining fully queryable.

• Expert technical support is included in our pricing, ensuring you receive the assistance you need whenever you need it. Learn how we're revolutionizing hosted database support at no extra cost.

You can try Timescale Cloud for free for 30 days to see if it’s right for you: sign up and see for yourself.