What information does Illumineto transfer (copy) from your corporate environment and what is done with it?

Illumineto Spark has a significant amount of API integrations.  For the purpose of this response, to illustrate this article will use one CRM system example (Salesforce) and one cloud storage example (Google Drive). The principles explained below are generally true for other CRM and cloud storage systems. 

1. Salesforce: All queries for Contacts, Leads, Opportunities, and Accounts are not cached. After a query, the user selects the appropriate record, and the record ID is stored in Spark.
2. Salesforce Content Libraries: If the user elects for Spark to do a federated search to include their Salesforce content libraries, Spark will index all PDF and Microsoft Office documents in selected content library folders. This index is stored in Spark's federated index, which is hosted by Microsoft Azure.  Each Spark customer's index is partitioned to ensure there is no "spillage" of content between customers. 
3. Google Drive: If the user elects for Spark to do a federated search to include their Google Drive, then Spark will cache an index similar to Salesforce.
4. Documents: Regardless of their source location (CRM or cloud storage), once a document is added to a Spark Page, that item is copied from its source location (e.g. Google Drive) and securely hosted in Illumineto's infrastructure. This is required so that the document can be viewed by the intended recipient of any Spark Page itself.