1) The data collection overhead (i.e., the overhead the agent adds to the application server) is less than 5%.
2) By Spring Insight (Open Source) I guess you refer to the dev edition. If so, Insight Ops is designed for production environments while the dev edition is for dev ops. As such the main difference is:
Insight Ops is shipped preconfigured to collect only 10% of the data (configurable) to impose less overhead on the container
3) You can monitor all JDBC queries regardless of the DB type and vendor.
3) I mean the following: can Spring Insight store metrics/jdbc queries/etc. to any SQL database like MySQL/PostgreSQL instead current H2 database?
If true, how I can generate a database schema for Insight.