docs: Adds documentation on third-party deps (#728)

Closes #716

docs/SUMMARY.md

@@ -5,6 +5,7 @@
   * [Introduction to Hydra](README.md#introduction-to-hydra)
   * [OAuth2 Case Study](README.md#oauth-20-case-study)
 * [Introduction Tutorial (5 minutes)](tutorial.md)
+* [Dependencies](environment.md)
 * Understanding Hydra
   * [OAuth2 & OpenID Connect](oauth2.md)
     * [Overview](oauth2.md#overview)

docs/environment.md

@@ -0,0 +1,17 @@
+# Dependencies & Environment
+
+ORY Hydra is built cloud native and implements [12factor](http://12factor.net) principles. The Docker Image is 5 MB light
+and thoroughly versioned with [verbose upgrade instructions](https://github.com/ory/hydra/blob/master/UPGRADE.md)
+and [detailed changelogs](https://github.com/ory/hydra/blob/master/CHANGELOG.md). Auto-scaling, migrations, health checks,
+it all works with zero additional work required. It is possible to run ORY Hydra on any platform, including but not limited
+to OSX, Linux, Windows, ARM, FreeBSD and more.
+
+ORY Hydra has two operational modes:
+
+* In-memory: This mode does not work with more than one instance ("cluster") and any state is lost after restarting the instance.
+* SQL: This mode works with more than one instance and state is not lost after restarts.
+
+The SQL adapter supports two DBMS: PostgreSQL 9.6+ and MySQL 5.7+. Please note that
+older MySQL versions have issues with ORY Hydra's database schema. For more information [go here](https://github.com/ory/hydra/issues/377).
+
+No further dependencies are required for a production-ready instance.