1. Architecture: Shinken is a monitoring framework written in Python that allows users to monitor their infrastructure via plugins and extensions, whereas Thanos is a monitoring tool designed for Prometheus metrics that provides a scalable, highly available solution for storing and querying Prometheus metrics.

2. Data Storage: Shinken stores monitoring data in its own backend database, which can be flexible but requires additional setup and configuration, while Thanos integrates directly with Prometheus and utilizes its storage system, simplifying the data storage process and ensuring compatibility with Prometheus metrics.

3. Scalability: Shinken's scalability depends on the resources of the server it is installed on and requires manual adjustments and configurations for scaling out, whereas Thanos is inherently highly scalable, utilizing distributed systems principles to handle large amounts of data without the need for extensive manual intervention.

4. Query Language: Shinken does not have a built-in query language for analyzing and visualizing monitoring data, requiring users to utilize external tools or plugins for this functionality, whereas Thanos comes with its own powerful query language that enables users to perform complex queries and analysis on Prometheus metrics data.

5. Alerting: Shinken offers basic alerting capabilities through its integration with Nagios plugins, allowing users to set up alerts based on threshold values, while Thanos lacks a native alerting system, relying on Prometheus Alertmanager for setting up and managing alerts based on monitoring data.

6. Community Support: Shinken has a smaller community compared to Thanos, which has a large and active user base contributing to its development and providing support through forums, documentation, and community resources.

In Summary, Shinken and Thanos differ in their architecture, data storage, scalability, query language, alerting capabilities, and community support in the monitoring and metrics space.