- Home
- Wide-Column Databases
- Baidu Tera
Baidu Tera
What it is
Baidu Tera is a free, self-hosted distributed NoSQL database for real-time applications. It stores sparse multidimensional tables organized by row key, column key, and timestamp, and the project describes scaling to petabytes across thousands of commodity servers. Its capabilities include strict consistency, MVCC, automatic failover, online schema changes, snapshots, and configurable sharding using ranged or hashed strategies. Storage options include RAMDISK, SSD, and disk tables, with tiered and block caches and Bloom Filters for real-time queries. Applications can connect through C++, Java, Python, and REST APIs. Tera’s architecture comprises an SDK, a master, and tablet servers; those servers handle read, write, and scan requests using an enhanced LevelDB storage engine. It can run on raw machines or in Docker, and a pseudo-distributed setup can run on one host using the local filesystem. The README describes use in Baidu products including web indexing, WebPage DB, and LinkBase DB for throughput-oriented and latency-sensitive workloads. The repository lists BSD-3-Clause and Apache-2.0 licenses and links to user manuals for pseudo-distributed mode, the API, and teracli.
For and against
For
- Free, self-hosted deployment
- Supports strict consistency and MVCC
- Offers automatic failover and snapshots
- Provides C++, Java, Python, and REST APIs
- Runs on raw machines or in Docker
Against
- Requires self-hosted deployment
- Listed client interfaces are C++, Java, Python, and REST
- Repository does not state security certifications or compliance claims
The verdict
Tera combines distributed table storage with consistency and operational features, and supports deployment from a single-host pseudo-distributed mode to larger clusters. Teams should account for its self-hosted deployment model and choose among its listed client interfaces.
Who it suits
Tera may suit teams seeking a self-hosted database for real-time workloads that need distributed tables, strong consistency, and horizontal scaling. Its listed clients include C++, Java, Python, and REST APIs.
On the record
- Architecture
- Tera consists of an SDK, a master, and tablet servers; tablet servers use an enhanced LevelDB storage engine and handle read, write, and scan requests.github.com · read 4 Oct 2026
- Client APIs
- The README lists C++, Java, Python, and RESTful APIs for accessing Tera.github.com · read 4 Oct 2026
- Client interfaces
- The project lists C++, Java, Python, and REST APIs.github.com · read 4 Oct 2026
- Consistency and availability
- Listed features include strict consistency, automatic failover, online schema changes, and snapshots.github.com · read 4 Oct 2026
- Data and consistency
- Tera supports MVCC, column-oriented storage with locality groups, and strict consistency.github.com · read 4 Oct 2026
- Data model
- Tera stores sparse, distributed, multidimensional tables indexed by row key, column key, and timestamp.github.com · read 4 Oct 2026
- Deployment
- The README says Tera can run on raw machines or in Docker, and describes a pseudo-distributed mode on a single host using the local filesystem.github.com · read 4 Oct 2026
- Documentation
- The English documentation index links to user manuals for pseudo-distributed mode, the API, and teracli, plus a performance benchmark page.github.com · read 4 Oct 2026
- Integrations
- Tera's README lists HDFS and other filesystems, ZooKeeper and Nexus lock services, and Sofa-pbrpc as infrastructure dependencies.github.com · read 4 Oct 2026
- Intended use
- The README describes use in throughput-oriented and latency-sensitive Baidu products, including web indexing, WebPage DB, and LinkBase DB.github.com · read 4 Oct 2026
- License
- The repository lists BSD-3-Clause and Apache-2.0 licenses.github.com · read 4 Oct 2026
- Operations
- Listed capabilities include automatic failover, online schema changes, and snapshots.github.com · read 4 Oct 2026
- Product
- Tera is a high-performance distributed NoSQL database designed for real-time applications.github.com · read 4 Oct 2026
- Scale
- The project says Tera can scale to petabytes of data across thousands of commodity servers.github.com · read 4 Oct 2026
- Security and compliance
- The repository lists BSD-3-Clause and Apache-2.0 licenses; the README does not state security certifications or compliance claims.github.com · read 4 Oct 2026
- Sharding
- Features include linear and modular scalability, automatic configurable sharding, and ranged or hashed sharding strategies.github.com · read 4 Oct 2026
- Storage and queries
- Tera supports RAMDISK, SSD, and disk tables, tiered cache, block cache, and Bloom Filters for real-time queries.github.com · read 4 Oct 2026
- Support and contribution
- The README welcomes contributions and provides [email protected] for joining the project.github.com · read 4 Oct 2026
