The Zscaler Digital Transformation study guides describe the Zero Trust Exchange using the conceptual model of “Brains and Engines.” Engines are the inline enforcement components—ZIA Public Service Edges, ZPA Service Edges, App Connectors, etc.—that sit in the data path to forward traffic, apply policy, and perform inspection.

The “Brains” side, however, represents the cloud control and intelligence plane. Here Zscaler hosts components such as Central Authority, policy and configuration stores, analytics engines, and, critically, the Logging and Reporting infrastructure (Nanolog clusters, Log Streaming Service, and analytics dashboards). The documentation explicitly associates log collection, compression, forwarding to SIEM/SOAR platforms, and long-term analytics with this centralized cloud layer rather than the enforcement engines themselves.

Engines generate rich telemetry, but they stream it back to the brains layer, where it is normalized, indexed, retained, and made searchable for investigations, compliance, and performance analysis. OneAPI is an access interface, not the location of the logging services, and “Memory” is not a formal architectural construct in the Zscaler model. Therefore, in the official architecture view taught for the exam, logging services clearly reside in the Brains component of the platform.

===========