February 2022 Summaries

In this month's newsletter, Timescale Cloud has released two new storage autoscaling features for multi-node services, allowing users to forget about full disks in their clusters. The company also announced a significant funding round of $110 million led by Tiger Global, valuing the company at over $1 billion. Additionally, TimescaleDB 2.6 was released with features such as compression and timezones, while Promscale 0.10.0 improved CPU performance and metrics instrumentation. The newsletter also featured new content on observability, including a Developer Q&A with Trading Strategy on building a data stack for crypto quant trading, and a tutorial on storing k6 metrics in TimescaleDB and visualizing them with Grafana. Furthermore, the company shared its plans to support and grow communities around the world, sharing educational resources as broadly as possible. The newsletter concluded by announcing new community features, including a forum, and virtual events, such as Office Hours and Community Day.

Building TimescaleDB on Docker involves several steps that can be simplified with the right guidance. To start, one needs to have Docker installed and set up on their machine. Then, a new container is created using a Docker image such as Ubuntu, which serves as the base for the TimescaleDB system. The container is then configured with necessary packages, including PostgreSQL, and the TimescaleDB repository is cloned from GitHub. The build process involves preparing the build system, building the extension, installing it, and setting up the PostgreSQL instance with TimescaleDB. This includes configuring the PostgreSQL configuration files, editing the "pg_hba.conf" file, and adding the TimescaleDB extension to the database. Finally, a Dockerfile can be created to automate the setup process, making it easier to spin up a TimescaleDB instance. By following these steps, users can build and run TimescaleDB on their machine, either for testing purposes or to contribute to the open-source codebase.

TimescaleDB 2.6 has been released with new features highly requested by the community, including compression for continuous aggregates and support for timezones in continuous aggregates. Compression allows users to save disk space by compressing data stored in a specific hypertable that is older than a certain period of time. The feature uses best-in-class compression algorithms along with a novel hybrid row/columnar storage design. Additionally, the new release includes an experimental feature called `time_bucket_ng`, which expands on the capabilities of the existing `time_bucket` hyperfunction by including features such as monthly buckets and timezones. This allows users to create continuous aggregates that can be used for downsampling and long-term analytics, while also reducing storage costs. The release is available now for Timescale Cloud and self-managed TimescaleDB instances.

Timescale has raised $110 million in its Series C funding, led by Tiger Global, alongside existing investors. This round brings the company's valuation over $1 billion and $180 million in total funding. Timescale has seen significant growth in its community and revenue, with 7x community growth and 20x revenue growth in just two years. The company is valued for its expertise in time-series data and its relational database, TimescaleDB, which is built on top of PostgreSQL. With this new funding, Timescale aims to accelerate its vision of building a data platform for developers, anchored around a best-in-class developer experience for PostgreSQL, time-series data, analytics, and data-driven applications. The company plans to invest in product, engineering, R&D, and serving its community and customers worldwide. This funding also enables Timescale to give back to the larger developer community, including building an entire team dedicated to upstream PostgreSQL contributions. With this new round of financing, Timescale is poised to continue innovating on top of PostgreSQL and delivering breakthrough capabilities that enable more developers to build exceptional data-driven applications.

This installment of the "Community Member Spotlight" series highlights Mikko Ohtamaa, CEO at Trading Strategy, who shares how TimescaleDB enables his company to provide traders and investors with direct access to high-quality trading strategies and real-time control over their assets. By integrating market data about thousands of crypto assets, algorithms, profitability simulation, and more into one solution, Trading Strategy can focus on solving business problems without building the infrastructure layer. Mikko Ohtamaa explains how TimescaleDB's powerful combination with PostgreSQL allows them to store large amounts of time-series data in a single database, making it simple for software developers to build on top of this, while also offering vertical scaling and continuous real-time aggregations. The company is completing its seed round and plans to launch user-accessible trading strategies later this year, as soon as they are confident the software stack is well-behaving and related smart contracts are secure.

The most frequent query applications make on time-series data is to find the most recent value for a given device or item. However, querying large time-series datasets can be challenging due to the impact of indexes and query patterns. Five methods are presented to access the most recent value in PostgreSQL: Naive GROUP BY, LATERAL JOIN, TimescaleDB SkipScan, Loose Index Scan, and Logging table with triggers. Each method has its advantages and disadvantages, and some may not perform well on large datasets or high-cardinality data. The choice of method depends on the specific use case and requirements of the application. It is recommended to test each option in an environment that closely resembles production to determine which approach works best for a particular use case. Additionally, using TimescaleDB can provide efficient querying capabilities, especially with SkipScan, and can help improve the performance of time-series data applications.