June 2020 Summaries

Team Timescale has released new features, documentation, and tutorials to help developers work with data. The team hosts virtual sessions and community meetups to share knowledge and resources. Recent releases include a new beta version of TimescaleDB multi-node, a Prometheus Connector alpha version, and a maintenance release of TimescaleDB 1.7.1. The team also shares tips and recommendations for using TimescaleDB with AWS and other tools, as well as stories from community members who have used the platform to analyze and visualize data. Additionally, the team promotes its community meetups and invites developers to join the discussion on Slack channels.

Ingest performance optimization for PostgreSQL and TimescaleDB is critical for applications that require fast data ingestion, such as application monitoring, analytics, IoT monitoring, and more. To improve ingest performance in vanilla PostgreSQL, consider using indexes in moderation, reconsidering foreign key constraints, avoiding unnecessary UNIQUE keys, using separate disks for WAL and data, and choosing performant disks. For TimescaleDB, use parallel writes, insert rows in batches, properly configure shared buffers, run Docker images on Linux hosts, write data in loose time order, avoid "too large" chunks, and watch row width to achieve better ingest performance. By implementing these techniques, users can improve the speed of ingesting new data into their databases.

Timescale stands with the Black community and against racism and hate. The company aims to be diverse, inclusive, courageous, and empathetic. In recent times, there has been a lot of racial violence and police brutality targeting the Black community in the US, which is unacceptable and tragic. Despite this, there are also hopeful signs as people from all races are rising up to demand change and some police departments have responded with compassion. The company recognizes that it has a responsibility to speak up and amplify the voices of marginalized communities. It encourages employees to use their privilege to benefit others and offers support for those who need it. Additionally, Timescale is exploring ways to be an agent of positive change against racism and injustice.

The DataPub meetup featured guest speakers Chris Whong and Jonathan Leek who shared their experiences solving "urban mysteries" using open datasets. Chris discussed how he uses PostgreSQL to clean and aggregate the New York City subway system's turnstile data, reducing 14M records to 2M, and making it possible to analyze trends and ask meaningful questions. He also highlighted the importance of open datasets and his work with Qri Cloud, an open source project for versioning and sharing datasets. Jonathan Leek, a Data Architect at the St. Louis Regional Data Alliance, presented on building a regional analytics database to make public data more accessible and usable in the St. Louis region, focusing on residential and commercial property data. His team is working on "The Regional Entity Database," an open source ETL infrastructure and analytical database that can be extended to all types of public data and municipalities. Both speakers emphasized the importance of making public data accessible and usable for citizens, organizations, and policymakers.

Grafana's alerting functionality allows users to configure visualizations and alerts for key metrics in the same place, making it a powerful tool for monitoring setups. The demo covers four parts: Alerting Principles, Alerts in Grafana, Let's Code: 3 Alerts and 3 Notification channels, and Alerts with NO DATA, covering topics such as universal best practices for alerting, how to define alerts using FOR and without FOR, and how to connect Grafana to popular notification channels like Slack, PagerDuty, and OpsGenie. The demo also provides resources and tutorials to help users recreate the sample monitoring setup and modify it to use their data sources and notification channels.

The TimescaleDB team has been busy releasing new features, creating tutorials and demos, and hosting virtual sessions to help developers work with time-series data. The team shares its favorite new resources in its biweekly newsletter, which includes technical content, videos, and tutorials on topics such as Grafana, Prometheus, and TimescaleDB. The team also highlights community spotlights, including a charitable foundation that uses open-source tools to monitor energy consumption and a project that aims to build a global environmental datastack for climate action. Additionally, the team announces its plans to make its multi-node version of TimescaleDB available for free, starting with a beta release.