Google Cloud Professional Cloud Architect Analyzing and Optimizing Technical and Business Processes Syllabus

Why is regular disaster recovery testing necessary, and what is a 'game day'?

Untested DR plans often fail when needed (stale runbooks, missing access, broken backups). A 'game day' is a scheduled exercise that deliberately injects failures or simulates a disaster to validate runbooks, RTO/RPO, and team readiness under realistic conditions.

Compare the four common DR architecture patterns by cost and recovery speed.

Backup & Restore: cheapest, slowest (restore from backups). Pilot Light: minimal core always running, scaled up on failover. Warm Standby: a scaled-down full copy running, ready to scale. Hot Standby / Multi-site Active-Active: full duplicate handling traffic, fastest (near-zero RTO/RPO) but most expensive.

What is an incident in the context of incident response, and what is an SEV/severity level?

An incident is an unplanned disruption or degradation of a service. A severity (SEV) level classifies its impact (e.g., SEV1 = critical, full outage / data loss; SEV2 = major degradation; SEV3 = minor) to drive the appropriate response, urgency, and escalation.

What is an escalation policy in incident response?

A predefined chain that defines who is paged, in what order, and after what time thresholds if an incident is unacknowledged or unresolved — ensuring incidents reach people with the right authority and expertise promptly.

What is the role of an Incident Commander (IC)?

The single person who coordinates the overall incident response — directing roles, making decisions, managing communication, and keeping the response organized — without necessarily performing the hands-on fixes themselves.

What is a blameless postmortem?

A retrospective after an incident that focuses on systemic causes and improvements rather than assigning individual blame, encouraging honest disclosure so the organization learns and prevents recurrence.

What is MTTR, and what do its variants MTTD and MTTA measure?

MTTR = Mean Time To Recover/Repair (avg time to restore service). MTTD = Mean Time To Detect (avg time to notice an incident). MTTA = Mean Time To Acknowledge (avg time for someone to respond to an alert).

What is change management in IT operations?

A structured process for proposing, reviewing, approving, scheduling, and documenting changes to systems to minimize risk and disruption, ensuring changes are controlled, traceable, and reversible.

Differentiate standard, normal, and emergency changes in ITIL change management.

Standard change: pre-approved, low-risk, routine (no CAB review needed). Normal change: requires assessment and approval (often via a Change Advisory Board). Emergency change: urgent fix for a critical issue, expedited approval, documented retroactively.

What is a Change Advisory Board (CAB)?

A group of stakeholders that reviews, assesses risk for, and approves or rejects proposed normal/significant changes, advising on prioritization and scheduling to protect production stability.

What is capacity management and what is its goal?

The process of ensuring IT resources (compute, storage, network) are sufficient to meet current and future demand at acceptable cost and performance — balancing under-provisioning (outages) against over-provisioning (wasted spend).

Distinguish demand management from capacity management.

Capacity management ensures supply of resources meets demand. Demand management influences or shapes the demand itself — e.g., via rate limiting, scheduling batch jobs off-peak, pricing/quotas, or smoothing usage spikes to fit available capacity.

What is the difference between vertical and horizontal scaling?

Vertical scaling (scale up) adds more power (CPU/RAM) to a single machine. Horizontal scaling (scale out) adds more machines/instances and distributes load across them. Horizontal scaling is generally more elastic and fault-tolerant.

What is a RACI matrix and what do its letters mean?

A responsibility-assignment chart mapping roles to tasks. R = Responsible (does the work), A = Accountable (ultimately answerable, one per task), C = Consulted (provides input), I = Informed (kept updated).

What is skill mapping (skills matrix) and why is it used?

A grid mapping team members against required skills/competency levels. It is used to identify capability gaps, plan training and hiring, assign work appropriately, and reduce single points of knowledge (bus factor).

What is stakeholder management, and what tool helps prioritize stakeholders?

The process of identifying, analyzing, and engaging people affected by or influencing a project. A power/interest grid (interest vs. influence matrix) helps prioritize: manage closely (high power/high interest), keep satisfied, keep informed, or monitor.

What is a communication plan and what key questions does it answer?

A document defining how project/operational information is shared. It answers: who needs information, what information, how often (frequency), through which channel/format, and who is responsible for delivering it.

What is the difference between governance and management in IT decision-making?

Governance sets direction, policies, and accountability — deciding what should be done and who decides (evaluate, direct, monitor). Management executes within that framework — planning, building, running, and monitoring day-to-day operations.

Name common organizational decision-making models, including consensus and consent.

Autocratic (single decider), consultative (decider seeks input), consensus (all must agree), consent (proceed unless someone has a reasoned objection), democratic/majority vote, and delegated (a designated owner decides). Consent is faster than consensus because it requires no objections rather than full agreement.

What is DevOps, in cultural terms?

A culture and set of practices that break down silos between development and operations, emphasizing shared ownership, automation, continuous delivery, fast feedback, and collaboration to deliver software reliably and quickly.

What is SRE (Site Reliability Engineering) and how does it relate to DevOps?

SRE is Google's prescriptive implementation of DevOps principles, applying software engineering to operations. It is often summarized as 'SRE implements DevOps' (class implements interface), adding concrete practices like SLOs, error budgets, and toil reduction.

Define SLI, SLO, and SLA.

SLI (Service Level Indicator): a measured metric of service behavior (e.g., latency, availability %). SLO (Service Level Objective): an internal target value/range for an SLI (e.g., 99.9% availability). SLA (Service Level Agreement): an external contract with customers including consequences/penalties for missing targets.

What is an error budget in SRE and how is it used?

The allowed amount of unreliability = 100% minus the SLO (e.g., a 99.9% SLO gives a 0.1% error budget). It is 'spent' by outages; while budget remains, teams can ship features freely, but if it is exhausted, releases freeze to focus on reliability.

What is 'toil' in SRE, and why reduce it?

Toil is manual, repetitive, automatable, tactical operational work that scales linearly with service size and provides no enduring value. SRE caps and reduces toil through automation so engineers can focus on engineering improvements.

In Google Cloud, what is a billing account and how does it relate to projects?

A billing account is a cloud-level resource that defines who pays and is linked to a Google payments profile. One billing account can be linked to many projects; every project that uses paid resources must be associated with exactly one active billing account.

What are budgets and budget alerts in Google Cloud Billing?

A budget sets a monetary spending threshold (for a billing account or scoped to projects/services). Budget alerts send notifications (e.g., email or Pub/Sub) when actual or forecasted spend crosses configured percentage thresholds (e.g., 50%, 90%, 100%). Note: alerts notify but do not automatically cap spending.

How can budget alerts be used to automatically cap or stop spending in Google Cloud?

By configuring the budget to publish alerts to a Pub/Sub topic, then triggering a Cloud Function that takes action — for example, disabling the billing account on the project — since budgets alone only notify and never auto-stop charges.

What are labels in Google Cloud and how do they support cost allocation?

Labels are key-value pairs attached to resources (e.g., env=prod, team=payments, cost-center=123). Exported to billing data (e.g., BigQuery), they let you break down and allocate costs by team, environment, application, or cost center.

What is the difference between labels and the project/folder hierarchy for cost allocation?

The resource hierarchy (Organization → Folders → Projects) groups resources structurally and is a natural cost-allocation boundary (per-project billing). Labels add flexible, cross-cutting cross-project dimensions (team, app, env) for finer-grained reporting that the hierarchy alone cannot express.

What is FinOps?

A cultural and operational practice (Cloud Financial Operations) that brings financial accountability to variable cloud spend, uniting finance, engineering, and business teams to make data-driven trade-offs between cost, speed, and quality. Its phases are Inform, Optimize, and Operate.

What is the difference between showback and chargeback in cloud cost management?

Showback reports each team/department its cloud usage and cost for visibility and awareness, without billing them. Chargeback actually allocates and bills those costs back to the consuming team's budget, creating direct financial accountability.

What is continuous process optimization, and what improvement cycle underpins it?

The ongoing practice of measuring, analyzing, and incrementally improving processes rather than one-time fixes. It is commonly driven by the PDCA (Plan-Do-Check-Act / Deming) cycle and Kaizen — using metrics and feedback loops to drive iterative improvement.