IBM Cloud Technical Advocate study plan

Define IaaS, PaaS, and SaaS in your own words, then place at least eight IBM Cloud offerings (VPC VSIs, OpenShift, Code Engine, Cloud Object Storage, Db2 on Cloud, IBM Cloud Databases) into the correct model and justify each placement.

Compare capital-expenditure and operating-expenditure models for a three-tier web application, listing which cost lines move from fixed to variable and which do not.

Explain the shared responsibility model three times: once for a VPC virtual server, once for OpenShift on IBM Cloud, and once for a fully managed database service, naming who patches the OS, the runtime, and the data in each case.

Build a business-value narrative for a skeptical CFO covering elasticity, time-to-market, disaster-recovery cost avoidance, and exit/portability risk, with at least one quantified assumption per claim.

Diagram the IBM Cloud resource hierarchy from account through resource groups, regions, zones, to individual service instances, and annotate where access control, billing, and fault isolation are each enforced.

Compare VPC virtual server profiles (balanced, compute, memory, very high memory), Bare Metal Servers, Power Virtual Server, and OpenShift or Kubernetes worker nodes on isolation, cost shape, and workload fit, then recommend one option for three given workloads.

Provision a VPC with two subnets in different zones, one virtual server instance in each, a load balancer with a health check, and a public gateway on only one subnet, then verify failover behavior by stopping one instance.

Explain when Code Engine beats OpenShift and when OpenShift beats a plain virtual server, using cold-start, scaling-to-zero, and operational-ownership arguments.

Select a storage type and tier for five scenarios covering a shared filesystem for containers, a high-IOPS transactional database, and archival compliance data, justifying each choice with IOPS, durability, and access-pattern reasoning.

Configure VPC network security using security groups, network ACLs, floating IPs, and public gateways, then describe a private-only workload path that reaches the internet without exposing any instance directly.

Create a Cloud Object Storage bucket with a lifecycle rule, a bucket policy, and a service credential scoped to a single bucket, then document how Vault, Cold Vault, and Smart Tier change the cost profile as object age increases.

Compare IBM Cloud Databases for PostgreSQL, Db2 on Cloud, and Databases for MongoDB on managed backups, high availability, and connection-secret handling, and pick one for a regulated workload.

Design an IAM model using access groups, service IDs, and trusted profiles, distinguishing platform management roles from service access roles and demonstrating why assigning policies to individuals does not scale.

Enable a customer-managed encryption key in Key Protect or Hyper Protect Crypto Services and bind it to an object storage bucket, then explain key rotation, key revocation impact, and the keep-your-own-key versus bring-your-own-key distinction.

Route Activity Tracker events to a persistent destination and identify the specific events produced by an IAM policy change, a key deletion, and a bucket access, explaining why each matters to an auditor.

Map one regulated workload to at least four IBM Cloud compliance certifications such as SOC 2 Type 2, ISO 27001, HIPAA, and PCI DSS, and state which controls remain the customer's responsibility under each.

Outline a three-step modernization path for a monolith using containerization, a Tekton or CI/CD pipeline, and observability, and name the point at which you would recommend re-platforming versus refactoring.

Deploy an application to Code Engine directly from source, confirm scale-to-zero behavior, and contrast the operational overhead against running the same container on OpenShift.

Invoke a Watson or watsonx.ai service endpoint to extract insight from sample text, and explain how prompt or model governance, data residency, and API key rotation apply in production.

Compare IBM Cloud support plans (Basic, Advanced, Premium) on response-time commitments and included advisory services, then select one for a production workload with a stated recovery-time objective.

Structure an account for cost attribution using resource groups and a mandatory tagging convention, then show how tags flow into billing and usage reporting.

Create a budget with spending alerts and forecast the cost of a two-zone workload using the Cost Estimator, documenting every assumption about hours, storage, and egress.

Apply enterprise governance by describing how an enterprise, account groups, and accounts relate, and where catalog restrictions and enterprise-level IAM policies override account-level settings.

Audit an existing account for idle or orphaned resources, unused floating IPs, and over-provisioned instances, and quantify the monthly savings of the clean-up.

Design an end-to-end landing zone for a regulated three-tier application covering VPC segmentation, private connectivity, key management, audit logging, and compute placement, and defend each choice against a stated compliance constraint.

Re-run two hands-on labs without instructions: one IAM policy binding with a service ID, and one VPC deployment with security groups and a load balancer, to confirm muscle memory rather than recall.

Rewrite your three weakest explanations from weeks one through six in plain language a customer would accept, targeting the specific points you got wrong on practice questions.

Trace a single user request from the public internet through the load balancer, security group, application instance, and database, naming every control and log source it touches.

Complete two full-length timed practice exams weighted by the published domain percentages, and score each domain separately to identify residual gaps.

Re-derive the business-value and cost-governance arguments from memory, since these are the highest-frequency reasoning questions and the easiest to under-prepare.

Review every missed item from weeks one through seven and rewrite each as a one-sentence rule you can apply to a new scenario rather than to the exact question seen.

Rehearse a five-minute verbal walkthrough of a full IBM Cloud solution covering architecture, security, cost, and support, without notes.