DC-SCM v1.2 (M1120) — Service Technician Training

A server is more than processors, memory, storage, and network interfaces. It's a managed platform that must be initialized, monitored, secured, updated, diagnosed, and recovered across its life cycle. Firmware — the low-level software stored on platform devices — does much of that work. The BIOS (Basic Input/Output System) initializes host hardware and starts the boot process, reporting POST (Power-On Self-Test) codes that show where initialization is succeeding or failing. The BMC (Baseboard Management Controller) monitors and controls the server independently of the host operating system.

Without a common module, every Host Processor Module (HPM) — the main logic board carrying program-specific compute, memory, accelerator, storage, and networking resources — would have to duplicate its own management and security circuitry, interfaces, firmware storage, service access, and trust functions. That duplication multiplies design effort, validation scope, platform variation, and service complexity.

The DC-SCM pulls those reusable management, security, monitoring, control, and service functions off the program-specific HPM. The two remain one server platform, joined by the DC-SCI interface, but the split lets many HPM designs share one standardized management and security architecture.

This course builds a practical mental model of DC-SCM v1.2 for hardware lab technicians. The sequence is deliberate:

Locate the function. HPM, DC-SCM, or the boundary between them.
Name the target. The exact controller, storage device, security element, and logical copy.
Trace the interaction. From the initiating tool through controller, interface, security boundary, and physical device.
Choose the supported method. The highest-level approved path that still reaches the target safely.
Verify the result. Not just that a write completed, but that the correct image is accepted, selected, active, and healthy.

Join this Course

Responsible	Administrator
Last Update	09/15/2026
Completion Time	7 hours 30 minutes
Members	0

Course

Phase I — Architecture First

4Lessons · 3 hr
- M1 — Platform architecture and the DC-SCI boundary
- M2 — DC-SCM component, storage, and BMC configuration layers
- M3 — Security processors, Roots of Trust, and terminology
- M4 — Dual-SPI physical and logical topology
Phase II — Interaction Tracing

5Lessons · 4 hr 30 min
- M5 — Trace method: transaction owner, control plane, and interface
- M6 — Managed update flow, reporting evidence, and connection authorization
- M7 — Component-specific methods and service configurations
- M8 — Authorized offline SPI programming and proof boundaries
- M9 — Recovery escalation, stop rules, records, and source precedence
Certification

1Lessons ·
- DC-SCM v1.2 (M1120) — Service Technician Certification

Architecture first • Trace every interaction • Verify the final state

This course builds the mental model a technician uses before touching a firmware tool: locate the function, name the target, trace the interaction, choose the supported method, and verify the result. Phase I builds the architecture model. Phase II traces interactions end to end and ends in a certification assessment.

SCOPE AND SAFETY BOUNDARY

This course teaches the architecture and the reasoning used to choose and verify a service method. It does not authorize a firmware operation or replace the current SKU-specific work instruction, release package, security policy, or rework procedure. Commands, package names, jumper settings, selectors, and recovery sequences vary by platform and release.

Audience: Platform and datacenter service technicians, lab recovery technicians, and technical support personnel who work with DC-SCM firmware and configuration.

Prerequisites: Basic firmware concepts, ESD and electrical safety, controlled-document use, and access to current platform-specific board maps, telemetry, approved packages, fixtures, and work instructions.