...
The Kyvos BI Server cluster is built processed on a Floating Master approach. Multiple instances of the Kyvos engine form a cluster governed by a coordination master. The coordination master is highly available. If there is a failure, its role is assumed by other nodes in the cluster without downtime. Requests are distributed to the appropriate Kyvos BI server using a round-robin algorithm.
...
Non‑browsing requests include dataset and semantic model build process requests, coordinating jobs between Kyvos and the “compute” engine (such as Spark or MapReduce), and user authentication requests. User authentication may be handled internally or delegated to another system like LDAP or Kerberos.
Query delegation handles requests from the Kyvos front end as well as from other applications such as Tableau, Excel, or Python. These application requests are translated into the native Kyvos query language and then passed to the individual Query Engines.
Once a query engine completes a query, the response is passed back to the BI Server. The BI Server consolidates the responses from the multiple query engines and then packages a single response back to the original application in the proper structure (Native Kyvos, MDX, or SQL).
...