Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

TigerGraph

tigergraph.com

  • Free plan
  • Linux
  • Web
  • API
  • Self-hosted
7.7 Club score Visit tigergraph.com

What TigerGraph is

TigerGraph is a graph database for analyzing connected data in real time. Its Native Parallel Graph design supports parallel computation and deep-link analytics, while vector similarity and hybrid graph-and-vector search extend the kinds of queries it can handle. Available graph query options include GSQL, OpenCypher, and GSQL pattern matching. Connectors cover Kafka, Amazon S3, Azure Blob Storage, Google Cloud Storage, Snowflake, BigQuery, PostgreSQL, Spark, Delta Lake, Iceberg, and Hudi.

Teams can install self-managed editions on Linux or as a Docker image; Enterprise Edition also supports Kubernetes clusters. TigerGraph's Savanna cloud service is available on AWS. Documented security measures include TLS, SAML 2.0 SSO, LDAP authentication, role-based access control, privileged-action audit logs, and encryption at rest. Community Edition is free, with a single-server limit and up to 300 GB of combined graph and vector data. Savanna is listed at $45/mo. A 30-day Enterprise Evaluation License is available unless another duration is agreed in writing.

Who it suits

TigerGraph suits teams that need real-time analysis across connected data, including graph and vector search. It offers self-managed deployment options as well as the AWS-based Savanna service.

Good to know, and what to watch

  • Supports parallel graph processing and deep-link analytics
  • Offers vector similarity and hybrid graph-vector search
  • Includes connectors for databases, cloud storage, and analytics tools
  • Documents access controls, encryption, and audit logs
  • Community Edition is limited to one server
  • Community Edition caps combined graph and vector data at 300 GB
  • Savanna is listed at $45/mo

TigerGraph Enterprise

—

Savanna Free Trial

Free

TigerGraph Community

Free

Savanna

$45/mo

Free plan

Getting started

  1. Open the TigerGraph website
  2. Choose Community Edition, Savanna Free Trial, Savanna, or another listed plan
  3. For self-managed use, install directly on Linux or run the Docker image
  4. Use Kubernetes clusters with Enterprise Edition when required
  5. Connect data sources such as Kafka, cloud storage, Snowflake, BigQuery, PostgreSQL, Spark, Delta Lake, Iceberg, or Hudi

Limits to check

Community Edition is limited to a single server and 300 GB of combined graph and vector data. The Savanna Free Trial provides one read/write workspace, two read-only workspaces, and manual backup; its Enterprise Evaluation License lasts 30 days unless another duration is agreed in writing.

The Club's review

TigerGraph is a graph database for real-time analysis of connected data, combining parallel graph computation with vector search and several deployment choices. It suits data and application teams working with relationship-heavy workloads, provided they can match the edition’s scale and operating model to their needs. The main decision is whether the Community limits are enough or whether managed capacity and enterprise controls justify the paid tiers.

Overview

TigerGraph uses a Native MPP Graph architecture and the GSQL query language to process connected data in parallel. That focus fits deep-link analysis, where useful answers require traversing multiple relationships rather than retrieving isolated records. ACID transactions and graph algorithms make it suitable for transactional production systems as well as analytical workloads.

The database also supports vector similarity search and hybrid graph-plus-vector search. This is useful when semantic retrieval must be combined with relationship context in one system, rather than maintained as separate graph and vector services.

TigerGraph was founded in 2012 and is headquartered in California. Its deployment range covers self-managed Linux or Docker installations, Kubernetes clusters with Enterprise Edition, and the Savanna managed service on AWS.

Key features

Queries and processing

Native MPP processing is designed for real-time, parallel deep-link analytics. Query choices include GSQL, OpenCypher and GSQL pattern matching, giving teams familiar alternatives while retaining TigerGraph’s native language. The combination is strongest for teams that need multi-hop graph analysis at operational speed; it is less compelling for applications that only need straightforward key-value or single-hop lookups.

Data connections

Connectors cover Kafka, Amazon S3, Azure Blob Storage, Google Cloud Storage, Snowflake, BigQuery, PostgreSQL, Spark, Delta Lake, Iceberg and Hudi. That breadth can reduce integration work when graph data must combine streaming events with warehouse and lakehouse sources.

Security and compliance

TigerGraph documents TLS, SAML 2.0 SSO, LDAP authentication, role-based access control, privileged-action audit logs and encryption at rest. Savanna has SOC 2 Type 2 audited compliance, while the on-premises product has PCI-DSS audited compliance. These controls make the platform more suitable for governed enterprise deployments than the free editions, which do not include eligible software support.

Pricing

TigerGraph uses a freemium model, with paid plans from $45/mo. Costs shown for Savanna are per GB/month for 24-7 service in the US/Tier 1 region.

Savanna Free Trial

The free trial costs nothing, billed at $0 per GB/month. Each workspace can use up to 32 vCPUs and 256GB RAM, with one read/write workspace and two read-only workspaces. Manual backup is included. The Enterprise Evaluation License lasts 30 days unless TigerGraph agrees in writing to a different term, making this tier appropriate for evaluation rather than an ongoing production service.

TigerGraph Community

Community is free and runs on a single server with up to 300 GB combined graph and vector storage. It is a practical starting point for individual developers and smaller deployments, but the single-server design limits resilience and growth. Community Edition is excluded from eligible software support.

Savanna

Savanna costs $45 a month, billed at $45 per GB/month. A workspace can reach 512 vCPUs and 4 TB RAM, with multiple read/write and read-only workspaces, automatic backups and rolling upgrades. This is the managed choice for teams that need more capacity and less infrastructure maintenance than Community provides.

Savanna Business Critical

Business Critical costs $126 a month, billed at $126 per GB/month, and adds multi-zone deployment, 99.95% uptime and premium support. Storage is $0.025 per GB/month. It fits services where availability and support justify a substantial premium over standard Savanna.

TigerGraph Enterprise

Enterprise uses custom pricing and adds high availability, disaster recovery, enterprise RBAC and security, rolling upgrades, scalability beyond 100 TB and standard support. It is the tier for large, regulated or mission-critical self-managed environments.

Platforms

TigerGraph supports API access, Linux, self-hosted deployment and the web. Self-managed software can run directly on Linux or as a Docker image; Enterprise Edition also supports Kubernetes clusters. Savanna currently runs on AWS, with regions in the United States, Europe, Asia Pacific and South America.

Who it's for

TigerGraph is a strong fit for organizations analyzing fraud, recommendations, networks or other connected datasets that need real-time multi-hop computation. It also suits teams combining graph relationships with vector retrieval and drawing data from several streaming, warehouse and lakehouse platforms.

Choose another approach if your workload is small enough for simple relational queries, if a single-server 300 GB ceiling is too restrictive, or if you do not need graph traversal and hybrid search. Savanna and Enterprise are aimed at teams able to fund and operate larger, governed deployments.

Pros and cons

  • Pros: Native parallel processing targets real-time deep-link analytics.
  • Pros: Graph, vector and hybrid search support can consolidate connected-data and semantic retrieval workloads.
  • Pros: Broad connectors cover major streaming, cloud-storage, warehouse and lakehouse systems.
  • Pros: Enterprise security controls and audited compliance support regulated deployments.
  • Cons: Community is limited to one server and 300 GB of combined graph and vector data.
  • Cons: Free tiers do not include eligible software support.
  • Cons: Business Critical pricing and per-GB billing can make larger managed deployments expensive.

Alternatives

For a wider comparison, see Graph Databases.

Sparksee is worth considering when cross-platform coverage, including mobile and desktop operating systems, matters and an academic free licence is relevant.

Memgraph is a better fit for teams seeking a free, open-source in-memory graph database with ACID transactions and on-disk persistence.

NebulaGraph suits users who want an Apache 2.0 open-source edition and community collaboration.

Aerospike Database is an alternative with a community edition capped at eight nodes, 2.5 TB of data and two namespaces.

Neo4j offers a free managed AuraDB plan and is self-hosted, making it a candidate for teams centered on that ecosystem.

Quine is an open-source option for readers prioritizing a free project.

TuGraph provides a free centralized community edition with high-availability deployment and TB-scale data.

InfiniteGraph has a free version capped at 50GB of disk storage.

Verdict

Choose TigerGraph when real-time, relationship-heavy analytics, hybrid graph and vector search, and broad data integration are central requirements. Its paid tiers provide a clear path from managed experimentation to highly available enterprise deployment. Look elsewhere if the 300 GB single-server ceiling is unacceptable, support must be included at no cost, or your application does not benefit from graph traversal.

The verdict

TigerGraph combines graph analytics, vector search, and a broad set of query and integration options. Check the Community Edition limits and deployment model against your requirements before choosing an edition.

Questions people ask

Is TigerGraph free?

Yes. Community Edition is listed as free, and Savanna has a free trial plan.

What does Savanna cost?

Savanna is listed at $45 a month. Savanna Business Critical is listed at $126 a month.

Which platforms are supported?

TigerGraph lists API, Linux, self-hosted, and web platforms. Self-managed deployment supports Linux and Docker, while Enterprise Edition supports Kubernetes clusters.

Does it support vector search?

Yes. TigerGraph supports vector similarity search and hybrid graph-plus-vector search.

What query languages are available?

Listed graph query options are GSQL, OpenCypher, and GSQL pattern matching.

What security controls are listed?

TigerGraph lists TLS, SAML 2.0 SSO, LDAP authentication, role-based access control, privileged-action audit logs, and encryption at rest.

What the maker publishes

Cloud deployment
Savanna is currently available on AWS, and its FAQ lists regions in the United States, Europe, Asia Pacific, and South America. source
Community limits
Community Edition is limited to a single server and up to 300 GB combined graph and vector data. source
Company background
TigerGraph says it was founded in 2012 and is based in Redwood City, California. source
Compliance
TigerGraph states that its Savanna service has SOC 2 Type 2 audited compliance and that its on-premises product has PCI-DSS audited compliance. source
Deployment
Self-managed TigerGraph can be installed directly on Linux or as a Docker image; Enterprise Edition also supports clusters through Kubernetes. source
Founded
2012 source
Graph and vector
The database supports vector similarity search and hybrid graph plus vector search. source
Graph processing
Its Native MPP Graph supports real-time deep-link analytics and parallel computation. source
Headquarters
Milpitas, California, United States source
Integrations
Connectors include Kafka, Amazon S3, Azure Blob Storage, Google Cloud Storage, Snowflake, BigQuery, PostgreSQL, Spark, Delta Lake, Iceberg, and Hudi. source
Purpose
TigerGraph is a graph database for real-time analytics on connected data, using Native Parallel Graph design and the GSQL query language. source
Query languages
TigerGraph editions list GSQL, OpenCypher, and GSQL pattern matching as graph query languages. source
Security
TigerGraph documents TLS encryption, SAML 2.0 SSO, LDAP authentication, role-based access control, privileged-action audit logs, and encryption at rest. source
Support
TigerGraph's support policy says standard support is included with a subscription; Community Edition and free-tier cloud services are excluded from eligible software support. source
Trial terms
The Enterprise Evaluation License lasts 30 days unless TigerGraph agrees in writing to another duration. source

Checked against the maker's pages on 30 Sept 2026.