November 2019 Summaries

TimescaleDB can be integrated with various AWS services to manage, store, and analyze time-series data at scale. Users have several options to set up their architecture, including Timescale Cloud, which provides a managed service offering on AWS, allowing for hands-off management and flexibility in customizing compute and storage configurations; EC2 instances, which offer granular control over the instance but require more operational responsibility; Kubernetes via Helm Charts, ideal for users with microservices architectures; Amazon Elastic Container Service (ECS), suitable for running TimescaleDB as a container to collect monitoring data from Prometheus; and AWS CloudWatch, Lambda, and TimescaleDB, which can be used to consolidate monitoring data in a single place. Each option has its benefits and drawbacks, making it essential for users to choose the one that best fits their needs and use case.

TimescaleDB 1.5 has been released with new features aimed at improving Total Cost of Ownership (TCO) by leveraging compression and data tiering functionality called move_chunk. Native compression allows users to store more data in less actual storage, resulting in an average compression ratio of 20x. Move_chunk enables users to move individual data chunks between Postgres tablespaces, reducing storage costs without needing to purge data. This flexibility is especially beneficial for environments where historical data analysis is a key part of the business. The feature also improves query performance by spreading read IO load across multiple storage mounts and disk arrays. Users can customize their setup based on their needs and follow simple steps to get started with move_chunk. The new release includes instructions, documentation, and a 30-day trial license for TimescaleDB Enterprise.