TigerData: can I provision and manage Tiger Cloud with Terraform/CLI, and where are the examples?

Most Tiger Cloud users want the same thing: provision databases the same way they manage the rest of their infrastructure—via Terraform, CLI, and automation—not by clicking around a UI. TigerData leans into that: you can fully manage Tiger Cloud programmatically, and you can even do it from your AI Assistant via Tiger MCP.

Below is a practical breakdown of what’s possible today, where to find examples, and how to wire it into a Terraform- and CLI‑driven workflow.

Quick Answer: Yes, you can provision and manage Tiger Cloud without touching the UI. TigerData provides a CLI (Tiger CLI) and an AI-integrated MCP layer that expose service management and SQL operations programmatically; Terraform-style workflows are supported by treating Tiger Cloud as API/CLI–driven infrastructure and wiring those commands into your IaC pipelines.

The Quick Overview

What It Is: A Postgres‑native cloud platform (Tiger Cloud) with a CLI and AI‑assisted control plane (Tiger MCP) that let you create, resize, and manage TimescaleDB services using commands and scripts instead of manual clicks.
Who It Is For: Teams who already treat databases as code—SREs, data engineers, platform engineers, and developers running telemetry, real‑time analytics, or AI workloads on Postgres.
Core Problem Solved: Removing “fragile and high‑maintenance” manual provisioning flows so you can manage Tiger Cloud services the same way you manage the rest of your infrastructure: reproducible, reviewable, and automated.

How It Works

From an operator’s perspective, there are three layers:

Tiger Cloud control plane – The managed Postgres + TimescaleDB service where each service hosts one database. This is where you get HA, automated backups, point‑in‑time recovery, and metrics.
Tiger CLI + Tiger MCP – A command‑line/AI control surface that lets you manage Tiger Cloud resources, execute SQL, inspect logs, and more, without using the web console.
Your automation (Terraform, CI/CD, scripts) – Terraform and pipelines orchestrate Tiger CLI/MCP commands so new environments, changes, and rollbacks are driven by versioned config.

In practice, you:

Install and authenticate the Tiger CLI.
Ensure the Tiger MCP server is active for your AI Assistant.
Use CLI or AI‑driven commands to:
- List, create, start, stop, resize, and fork services.
- Manage passwords, view service details, and collect logs.
- Execute SQL migrations and manage TimescaleDB features (CREATE HYPERTABLE, compression policies, continuous aggregates, etc.).
Wrap those commands in Terraform null_resource/local-exec, Makefiles, or CI workflows for full IaC.

1. Install and configure Tiger CLI

You start by installing the CLI binary and configuring secure authentication:

# Example: download, install, and authenticate (pattern)
curl -sSL https://download.tigerdata.com/cli/install.sh | bash

tiger login   # opens browser or prompts for token, depending on platform

Note: Exact install commands are in the TigerData docs under “Tiger CLI” and “Tiger MCP” setup. Always follow the latest instructions for your OS and shell.

Once configured, the CLI is your entry point for managing Tiger Cloud services via commands or scripts.

2. Check that Tiger MCP is active

Tiger MCP is the machine control protocol that ties your AI Assistant into Tiger Cloud. When it’s enabled, your assistant can call “Tiger tools” directly.

From the docs, you’ll see guidance like:

> is the tigerdata mcp server active for you?
⏺ Yes, the Tiger MCP server is active! I have access to a comprehensive set of Tiger Cloud (Timescale) tools, including:

Service Management:
  - List, create, fork, start, stop, and resize database services
  - View service details and logs
  - Update passwords

Database Operations:
  - Execute SQL queries against ...

If you see this kind of response, your AI Assistant can already:

Provision services
Resize compute
Fork databases for testing
Run SQL migrations
Pull logs and metrics

3. Use service management commands

Tiger MCP exposes a set of tools that the CLI and your AI Assistant can call. Conceptually, the operations look like:

service-list – enumerate all services
service-create – create a new Tiger Cloud service
service-resize – scale compute and storage
service-start / service-stop
service-fork – clone a production instance to staging/test
service-logs – tail or fetch logs

In a Terraform or CI context, you call these through the CLI or via MCP‑aware automation, for example:

# Create a new service with a defined plan/size
tiger services create \
  --name my-telemetry-prod \
  --plan performance \
  --region us-east-1 \
  --pg-version 16

# Resize when load grows
tiger services resize \
  --name my-telemetry-prod \
  --compute 4vCPU \
  --storage 2TB

Important: The exact flags and subcommands may differ; always consult the current tiger --help and online docs. The pattern, however—services create, services resize, etc.—is stable.

4. Run SQL and manage TimescaleDB features

Once a service exists, you can manage schema, indexes, hypertables, compression, and continuous aggregates from the CLI or AI Assistant:

# Run a migration file
tiger sql exec \
  --service my-telemetry-prod \
  --file migrations/001_create_hypertables.sql

Example SQL inside that file:

-- Turn a plain table into a hypertable
SELECT create_hypertable('metrics', by_range('time'), if_not_exists => TRUE);

-- Add space dimension for tenant_id or device_id
SELECT add_dimension('metrics', 'device_id', number_partitions => 32);

-- Configure compression for older chunks
ALTER TABLE metrics
SET (
  timescaledb.compress,
  timescaledb.compress_orderby = 'time DESC',
  timescaledb.compress_segmentby = 'device_id'
);

SELECT add_compression_policy('metrics', INTERVAL '7 days');

That’s the typical TigerData pattern: use Postgres SQL, plus TimescaleDB primitives, managed via CLI instead of manual psql sessions.

Features & Benefits Breakdown

Core Feature	What It Does	Primary Benefit
Tiger CLI & MCP tools	Expose service lifecycle and SQL operations as commands/tools.	Treat Tiger Cloud like any other API‑driven infrastructure.
Service management commands	Create, resize, fork, start/stop, and inspect services programmatically.	Reproducible environments, easy blue/green and staging setups.
Postgres‑native SQL control	Run migrations, configure hypertables, compression, and retention via SQL.	Keep everything in SQL—no separate DSL or custom query engine.

Ideal Use Cases

Best for GitOps / Terraform pipelines: Because you can wrap Tiger CLI commands inside Terraform null_resource or CI jobs, apply changes from PRs, and keep all stateful operations auditable and repeatable.
Best for multi‑environment telemetry stacks: Because you can script service creation and forking (prod → staging) to test new TimescaleDB features, compression policies, or schema changes against realistic data without manual cloning.

Where to Find Examples

Today, TigerData’s example patterns live primarily in:

Tiger CLI / MCP documentation – Step‑by‑step guides for:
- Installing Tiger CLI
- Setting up secure authentication for Tiger MCP
- Managing Tiger Cloud resources via your AI Assistant
Command help and examples – tiger --help and subcommand help typically show one‑line usage examples for service management and SQL execution.
GitHub and docs examples – The docs have code snippets showing:
- How to execute SQL against your Tiger Cloud service
- How to configure hypertables, compression, and continuous aggregates
- How to integrate Tiger Cloud metrics into external monitoring

Note: If you’re specifically looking for a Terraform provider: treat the current pattern as “Terraform drives Tiger CLI/MCP,” not “Terraform talks directly to Tiger Cloud via a first‑class provider.” The IaC examples usually show CLI‑driven workflows that are easy to wrap inside Terraform or other orchestration tools.

A common Terraform pattern looks like:

resource "null_resource" "tiger_service" {
  provisioner "local-exec" {
    command = <<EOT
      tiger services create \
        --name ${var.service_name} \
        --plan ${var.plan} \
        --region ${var.region}
    EOT
  }
}

You then add:

SQL migrations via local-exec or a CI migration job.
Output variables populated from tiger services describe JSON.

Limitations & Considerations

No “magic” Postgres provider: Tiger Cloud is Postgres + TimescaleDB, but you still manage it via Tiger CLI/MCP; Terraform doesn’t directly speak to the Tiger control plane unless you wrap CLI calls.
State vs. side‑effects: Terraform is declarative; CLI commands are imperative. It’s important to design your workflow so that service names and IDs are stable and your apply actions are idempotent. In many setups, you:
- Use Terraform to create networking and application resources.
- Use CI (triggered by Git changes) to run Tiger CLI commands that manage the database lifecycle and migrations.

Pricing & Plans

Tiger Cloud pricing is designed to be predictable and transparent:

You pay for the service plan (Performance, Scale, Enterprise tiers) and the compute/storage you allocate.
You do not pay per query or per backup:
- No per‑query fees.
- No extra charges for automated backups.
- No hidden ingest/egress networking fees for normal usage.

Typical plan framing:

Performance: Best for teams needing fast Postgres for telemetry and analytics, with managed HA, automated backups, and TimescaleDB features out of the box.
Enterprise: Best for organizations needing HIPAA, SOC 2 Type II, advanced networking (VPC peering/Transit Gateway), 24/7 support, and strict SLAs.

Important: For exact pricing, available regions, and plan details, always refer to the TigerData pricing page or Tiger Console—these can change as new regions and capabilities roll out.

Frequently Asked Questions

Can I manage Tiger Cloud purely via Terraform without using the Tiger CLI?

Short Answer: Not directly; you currently manage Tiger Cloud via Tiger CLI/MCP and integrate that into Terraform or CI workflows.

Details: TigerData exposes service management and SQL operations through the CLI and Tiger MCP. Terraform doesn’t yet have a dedicated Tiger provider that talks to the control plane natively. The recommended pattern is:

Use Terraform for your infrastructure (VPCs, subnets, Kubernetes, app services).
Use null_resource or CI pipelines to call tiger commands for:
- Creating and resizing services
- Managing passwords
- Running migrations and TimescaleDB configuration

This keeps your database lifecycle under version control while using the supported, documented Tiger Cloud interface.

Can my AI Assistant actually create and manage Tiger Cloud services for me?

Short Answer: Yes, if Tiger MCP is active and configured, your AI Assistant can list, create, fork, resize, and manage Tiger Cloud services.

Details: When the Tiger MCP server is active, your assistant has access to a toolkit that includes:

Service management: list/create/fork/start/stop/resize services, view details and logs, update passwords.
Database operations: execute SQL queries against Tiger Cloud services.

You can literally type instructions like:

“Create a new Tiger Cloud service named metrics-staging based on the same configuration as metrics-prod, then run this migration script”

Your assistant will translate that into the correct CLI/MCP calls. This is especially powerful combined with Git‑based workflows: reviewers see both the IaC diff and the migration scripts, while MCP handles the low‑level commands.

Summary

Tiger Cloud doesn’t force you into click‑ops. You get a Postgres‑native, TimescaleDB‑powered service that you can fully manage via CLI and AI‑driven MCP tools: create and resize services, run SQL, configure hypertables and compression, and inspect logs—all from scripts, CI, or your AI Assistant. Terraform‑style workflows are supported by wrapping Tiger CLI/MCP operations in your existing infrastructure pipelines, so your telemetry and analytics databases live in the same lifecycle as your apps.

Next Step

Get Started