1. Data Model: OpenTSDB is a time series database that stores large amounts of timestamped data. It uses a schema-less approach where data points are stored as key-value pairs. On the contrary, RavenDB is a document database that stores data in a JSON format, allowing for more flexible and complex data structures.

2. Query Language: OpenTSDB provides a powerful query language specifically designed for time series data. It offers functionalities like aggregation, filtering, and down sampling. In contrast, RavenDB uses LINQ (Language Integrated Query) for querying data, which allows for more expressive and intuitive querying capabilities.

3. Consistency Model: OpenTSDB uses a eventually consistent model as it replicates data across multiple nodes but may have delays in synchronization. On the other hand, RavenDB supports ACID transactions and offers strong consistency to ensure data integrity and reliability.

4. Clustering and Scalability: OpenTSDB is designed to be horizontally scalable and can handle massive amounts of data through distributed clusters. RavenDB also supports clustering for scalability but focuses more on vertical scaling by allowing data to be stored on multiple nodes within a single cluster.

5. Community and Support: OpenTSDB has a strong community base with active contributors and community support for troubleshooting and development. In comparison, RavenDB offers commercial support options along with a vibrant community for assistance and resources.

6. Use Cases: OpenTSDB is primarily suited for applications dealing with large-scale time series data such as IoT sensors, monitoring systems, and financial data analysis. RavenDB, on the other hand, is well-suited for applications requiring flexible schema design, document storage, and ACID compliance for transactional data.

In Summary, OpenTSDB and RavenDB differ in terms of their data models, query languages, consistency models, scalability options, community support, and ideal use cases.