TimescaleDB (by Tiger Data)

🐯 Timescale is now TigerData! 🚀 Eight years ago, we started as a PostgreSQL-based time-series database. But innovation never stands still—and neither have we. Today, we proudly unveil @TigerDatabase, marking our evolution into the fastest and most powerful PostgreSQL database, purpose-built for modern transactional, analytical, and AI-driven workloads. 🌟 Why TigerData? We’ve grown far beyond our initial identity. From handling early renewable energy use-cases to powering Bloomberg’s critical geotemporal data, our customers showed us we're not just the best in time-series—we're the best in PostgreSQL, period. 🔑 Our Journey Highlights: - 2017: TimescaleDB introduces groundbreaking hypertables. - 2019: Native columnar compression; our cloud service launches. - Recent innovations: 2500x faster distinct queries, 8x more efficient Boolean storage, and ultra-fast queries on high-cardinality data. 🚀 Customer Impact: - Bloomberg leveraged us for massive geotemporal datasets. - Renewable energy utilities replaced Redis to efficiently monitor operations. 🤝 Community Commitment: Our core remains open-source, including the beloved TimescaleDB extension. This rebrand changes nothing about our dedication to our community and customers. 🌐 Looking Ahead: Exciting innovations await: - Deeper lakehouse integrations. - AI-native features. - Enhanced scalability and unmatched performance. We are not just renaming; we're recommitting—to deliver on our mission of building the fastest Postgres, and to give our customers speed without sacrifice. ✨ Join the Tiger Era! 🐯 Your feedback is invaluable. Tell us your thoughts and what you'd love to see next. Ready for the ride? Let's roar ahead together! tsdb.co/tigerdata

Forty looks good on Postgres. Turns out time scales: • Still ridiculously reliable • Doesn't chase every new trend • Decades of runtime experience • Somehow keeps getting better That's not a database. That's the engineer you actually want on call. tsdb.co/wo5r8edl

Up to 160x faster UPDATE/DELETE on compressed data in TimescaleDB 2.27. Bloom filters prune batches before decompression. Also: smarter UPSERT pruning, broader vectorized execution in Hypercore, and refresh + compress in one aggregate policy. tsdb.co/tsdb-227-x

Great to see a full room for @michaelfreedman's session, From Asset to Fleet: Solar and BESS Operations with Ignition and TimescaleDB, at Solar & BESS Discovery Day. As solar and battery storage operations scale, so do the data challenges. Mike shared how teams are building scalable data architectures to move from asset-level monitoring to fleet-wide visibility and real-time analytics. Thanks to @InductiveAuto for hosting a great event and everyone who joined the discussion!

Most industrial orgs are drowning in data they can't actually analyze in real-time. Traditional databases + legacy historians = architecture that worked in 2015, not today. That's why modern time-series databases matter. Stay operational, stay analytical, stay on Postgres. No splits. No lag. Just data that works. linkedin.com/posts/timescal…

Intelligence isn't the bottleneck in robotics. Operational data infrastructure is. Robots can reason. But reliability at scale requires learning from millions of real-world cycles. That data problem has no internet shortcut. Click to read: automationworld.com/factory/roboti…

Recording of TJ's great presentation about pg_textsearch's design and implementation at recent pgconf dev. Link to Youtube video and GitHub repo below.

The order is the story: each wall creates the next. Full breakdown with the data: tsdb.co/five-walls-x

Water: gets the worst headlines, has the most engineering already fixing it. Closed-loop cooling can cut usage by up to 95%.

AI capacity problems get treated as five separate crises: GPUs, memory, cooling, power, water. They're one problem, and they hit in a fixed order.

Real-time analytics can unlock new ways to monitor complex systems. This case study by @TigerDatabase explores how data compression helped enable real-time energy tracking at scale: hackernoon.com/how-float-runs…

The clearest signal from ten events this spring: fewer people asking "what is a time-series database" and more asking about compression ratios and Hypercore under high-cardinality dashboards. The market moved past category education. tsdb.co/spring-events-x #Postgres #TimeSeries

One governed TimescaleDB table. Three protocols (Modbus, OPC UA, MQTT). Zero middleware. Claude Code Agent Teams build the collection layer and enforce the UNS naming contract while the code is written, not reconciled after. Walkthrough + failure path: tsdb.co/plant-floor-cl…

"@TimescaleDB is the new standard." That's @Siemens' conclusion for their industrial data historian. Replacing their use of Oracle, SQL Server, InfluxDB, and vanilla Postgres. Powering their next-generation industrial data platform. 🤌

25 years in building automation. Zero production TypeScript. Now runs a SaaS rendering 12 months of energy data across 38 buildings in under a second. Telemetry + materialized views + pgvector, all in one Postgres. Full community spotlight: tsdb.co/apexanalytica-x

10,000 GitHub contributions for @TimescaleDB! That's a metric to celebrate! 🥳

⭐️ If TimescaleDB has been useful to you, star us on GitHub: github.com/timescale/time…

TimescaleDB just hit 10,000 contributions. Every ticket, comment, and PR got us here, from Tiger Data engineers to contributors around the world. Thank you. On to 100,000. #TimescaleDB #PostgreSQL #OpenSource