TL; DR: Timescale delivers a well-rounded PostgreSQL platform for efficient database management. The company has incorporated unique functionality and features into its platform to enable developers to save time and data storage costs on their development projects. Dev teams can easily scale their databases without affecting performance and analyze their data with continuous aggregation. In the summer of 2023, Timescale plans to launch a data tiering service, enabling customers to automatically tier older data to more affordable object storage and maintain access to that data.
Every entrepreneur begins their journey with an idea. And that idea usually spawns from a pain point or a gap in the market that needs a much-needed market solution. Sometimes these solutions come from within an existing company that solves a problem for itself and looks to share this solution with the general public. The latter instance can easily describe the road that the team at Timescale took for its company.
Before Timescale launched, founders Mike Freedman and Ajay Kulkarni ran a company called Iobeam. They created Iobeam to provide a full-stack offering to connect IoT systems to cloud data services. It also offered client software development kits and server-side software for collecting client data. But the team couldn’t find a cloud data storage solution that fit client data needs.
“The developers all loved the PostgreSQL database, but it couldn’t natively handle the volume and unique needs of time-series data very well,” said Bryan Clark, Director of Products at Timescale. “So, a small team set about creating a better way to store time-series data in a PostgreSQL database, and from that, Timescale was born.”
Timescale offers a PostgreSQL cloud platform for time-series and relational database applications. With the rise of data collection, Timescale helps developers build applications with scalable databases for efficient processing and storage. Its top-notch performance and cost-effectiveness separate Timescale from other database competitors.
“Timescale offers storage compression, better performance, and aggregate data features—all built into a rock-solid PostgreSQL offering—that balances price, performance, and maintenance in favor of developers,” said Bryan.
Balancing Price, Performance, and Maintenance
Efficient database management requires balance. And balance is something Timescale knows a lot about and prioritizes in its operational and product framework. The team set out to build a platform that could balance price, performance, and maintenance, making the developer experience smoother and more effective in multiple ways.
Users can experience this balance first-hand with Timescale’s database makeup and features. As applications scale and gather more data, storing a relentless stream of events in a database table becomes an issue. Although most databases can handle massive amounts of data, query performance dwindles as data increases.
“Timescale developed a feature called hypertables, which automatically partitions the data to maintain higher query and ingest performance. While hypertables can reduce query planning times by 100x or more compared to vanilla PostgreSQL for the same number of partitions, the cost of storing this volume of data is a problem,” said Bryan.
But the Timescale team also accounted for this expense issue and leveraged data compression to reduce user storage costs. Timescale uses columnar storage and type-specific compression algorithms to save space within databases. Bryan said by integrating compression into its hypertables, developers could see 170 GB of storage reduced to only 10 GB.
As for maintaining databases, developers with large datasets usually want quick access to aggregate information. With the Timescale platform built on PostgreSQL, users don’t have to learn new languages because SQL queries allow access to all their data. Timescale also developed additional functions, including time_weight. These functions enable teams to create aggregate rollups for any timescale, from a day to weeks and years.
“Continuous aggregates create materialized views, so this rollup information is precalculated in the background—and we’ve seen queries that are thousands of times faster, and even tens of thousands of times faster.
Fostering a Forward-Looking Culture and Road Map
Timescale is an innovative company, and this defining forward-thinking culture started with its founders. Bryan told us one of Timescale’s greatest achievements was building a great company with a team culture of curiosity and constant improvement.
“Our constant improvement culture is exemplified in how the founders, Mike and Ajay, have reinvented the company a number of times, and this reinvention ensures we are always forward-looking,” said Bryan.
Much of Timescale’s success has stemmed from these values, leading to the creation of many of its cutting-edge features, including compression, continuous aggregates, and data tiering outside the database. But the influence on its road map innovation and development also comes from other sources.
Bryan said the use of market trends is limited because it is often a lagging metric for the team. Since market trends tell you where the market has already gone, companies usually have to play catch-up, and often it’s too late. Although market trends don’t play much of a role in its development endeavors, the Timescale team gathers and listens to user feedback from its customers and the open-source community to promote new feature ideas.
“We also look at new technologies, like advances in compression, which could give Timescale and our customers an edge. We combine all this to determine the Timescale road map,” said Bryan.
What’s Next: Data Tiering for More Cost-Effectiveness
Timescale has helped tons of users and big names, including Marvel Studios and Apple, store and manage their time-series databases. Time-series data can be found anywhere, from financial institutions to music groups, and Timescale powers more than 3 million active databases and counting. Timescale’s functionality has allowed companies to accelerate development time and better analyze their data with continuous aggregates.
But the Timescale team isn’t done advancing its platform. Led by its innovative culture, the team continues to evolve its features to meet new challenges and additional needs. Bryan told us that Timescale has a number of exciting things in the pipeline in 2023. He highlighted the company’s new data tiering service that is already available for early access but will be open to all in the summer.
“Data Tiering allows customers to automatically tier older data to a lower-cost bottomless object storage, which is still freely queryable via their database,” said Bryan.
With this feature, Timescale extends its flexibility capabilities and cost-effectiveness for better price performance. Users can still leverage the older data for their applications but at a fraction of the cost of regular database storage. As for Timescale’s other releases, the team is keeping those under wraps for now. “And without giving much more away, I’ll just say to be on the lookout for more ways Timescale will enable developers to build and manage data-intensive apps,” said Bryan.