What is Tiger Data (tigerdata.com) and what does it offer?

Updated

Tiger Data is a PostgreSQL platform built by the creators of TimescaleDB, aimed at time-series and machine data at petabyte scale. It is a fit if you need to ingest high-frequency sensor or machine data, keep years of history, and query live and historical data together in one Postgres — and you want to evaluate it through a free trial or a sales demo before committing.

Who it's for

Tiger Data positions itself around "sensor and machine data" — factories, wells, grids, and fleets that produce continuous streams of readings. The site cites a single TimescaleDB instance handling 3 trillion metrics per day, 1 quadrillion data points stored, and 3 petabytes of data volume. That scale claim is the core of the pitch: one Postgres instance rather than a separate time-series database plus a relational store.

If your workload is ordinary transactional data, the time-series features below add little. If your workload is high-volume, append-heavy metrics with long retention and mixed live/historical queries, the feature set maps directly to those needs.

Core capabilities

Capability What it does Why it matters for machine data
Automatic partitioning Time- and key-based partitioning Keeps reads and writes fast as tables grow into billions of rows
Row-columnar storage Row storage for writes, columnar for analytics, with compression Fast ingest without sacrificing analytical query speed
Tiered storage Hot data on SSD, colder data on low-cost object storage Store years of history without paying SSD prices for all of it
Lakehouse integration Ingest from Kafka and S3, replicate to Iceberg Fits existing streaming and data-lake pipelines instead of replacing them
Time-series functions 200+ SQL functions for time-based analytics Gap filling, downsampling, and time-bucketed aggregates without custom code
Interface SQL, APIs, CLI, and UI, Postgres-native Existing Postgres tooling and skills carry over
Search Hybrid retrieval with keywords, vectors, filters, ranking Useful when machine data is queried alongside text or embeddings

The design theme is consolidation: ingestion, compression, retention tiering, and analytics all live inside Postgres, so you avoid running a separate time-series system next to your relational one.

What users report

The site quotes Martin Zemko, a software engineer at CERN, describing speed-ups "from 10x to 40x, depending on the query, range, and data frequency," and calling Hypercore "a big win." Treat these as vendor-published customer statements rather than independent benchmarks — useful as a signal of the workload types where gains appear (large time ranges, high-frequency data), not as a guarantee for your queries.

Operational and compliance posture

For production adoption, the site lists:

  • Security: encryption at rest and in transit, private networking, access controls
  • Reliability: high availability, automated backups, point-in-time recovery
  • Compliance: SOC 2 Type II and GDPR support

These are the checkboxes most enterprise reviews ask for; verify specifics against your own requirements during a demo rather than relying on the summary.

How to evaluate it

  1. Start a free trial if you want to test ingestion and query performance on your own data shape. The site offers this as the primary self-serve path.
  2. Request a demo or contact sales if you need to discuss migration from an existing Postgres or time-series setup, or want the compliance and reliability details confirmed for procurement.
  3. Check pricing on the pricing page before assuming cost structure — the site links to it directly, and plan details are not summarized here.
  4. Test the migration path if you already run Postgres: the pitch is that TimescaleDB-style capabilities extend standard Postgres, so a proof-of-concept on a representative table is the fastest way to confirm fit.

Common sticking points to plan for

  • Data volume assumptions: the petabyte-scale claims describe large deployments; validate performance at your actual ingest rate and retention window during the trial.
  • Lakehouse dependencies: Kafka, S3, and Iceberg integration assumes those systems exist in your stack — confirm the connectors match your versions.
  • Compression trade-offs: columnar compression favors analytical reads; check that your write pattern and query mix benefit rather than adding complexity.

Bottom line

Tiger Data is worth a trial if you run high-volume time-series or machine data on or near Postgres and want partitioning, compression, tiered storage, and time-series SQL in one system. If your data is small, purely transactional, or already well served by a separate analytics stack, the value is lower. Use the free trial for hands-on validation and sales contact for migration, compliance, and pricing specifics.