IBM Systems and Solutions Architect

Good designs come from separating what a system does from how well it must do it. You turn requirements into design, weigh quality attributes against each other, and record decisions so the reasons survive.

• Functional and non-functional requirements
• Quality attributes: availability, scalability, performance, security
• Common architectural styles and their trade-offs
• Modeling with UML and the C4 approach
• Making and recording decisions
• Reviewing a design against its requirements

You build A system design with diagrams and three recorded architecture decisions.

Most architecture problems start as requirements problems. You analyze stakeholders, gather and prioritize requirements by several methods, plan communication, and control change once work is underway.

• Stakeholder analysis by influence and interest
• Eliciting requirements: interviews and workshops
• Writing requirements and user stories clearly
• Prioritizing what to build first
• Traceability and change control
• Communication planning
• Project management basics for architects

You build A stakeholder map and a prioritized requirements list for a real project.

Systems exist to support how a business works, so you need to see that work clearly. You map processes in BPMN, analyze where they waste time, and propose improvements using Lean and Six Sigma ideas.

• Process mapping and its purposes
• BPMN elements and notation
• Modeling an as-is process
• Finding bottlenecks and waste
• Lean and Six Sigma thinking
• Designing a to-be process

You build An as-is and to-be process model in BPMN with the improvements justified.

Modern systems are built to run on cloud platforms. You learn the principles of cloud-native design, how microservices and containers fit together, and how DevOps and agile practice change what an architect specifies.

• Principles of cloud-native design
• Microservices and their boundaries
• Containers and orchestration with Kubernetes
• Twelve-factor ideas for applications
• Continuous delivery and DevOps
• Choosing cloud services against building your own

You build A cloud-native architecture proposal for one business application.

Data outlives the applications that create it. You compare transactional and analytical systems, model data for each, choose between relational products such as MySQL and PostgreSQL, and design the data layer of a system.

• Transactional and analytical workloads
• Relational modeling and normalization
• Data warehouses and dimensional models
• Choosing between database products
• Data governance and ownership
• Designing the data architecture of a system

You build A data architecture with a transactional model, an analytical model and the reasons for each.

Systems must share data, and legacy data must move. You design integration with pipelines, APIs and messaging, plan storage including data lakes, and plan a migration that can be tested and reversed.

• Integration patterns: batch, API and messaging
• ETL and ELT pipelines
• Data lakes and storage tiers
• Assessing and profiling legacy data
• Migration approaches: big bang and phased
• Validation, cutover and rollback

You build A data migration plan from a legacy system into a data lake, with validation steps.

A design is only complete when it can be tested, released and kept alive. You plan layered testing, choose a deployment approach, and set up for maintenance, monitoring and the control of technical debt.

• Test levels and test planning
• Test environments and test data
• Deployment strategies and release planning
• Monitoring and operations handover
• Types of maintenance
• Managing technical debt

You build A test, release and maintenance plan for the system designed earlier in the course.

Security cannot be added at the end. You assess risk, model threats, choose safeguards for networks, identities and data, and plan the testing that shows whether the controls work.

• Risk assessment for a system
• Threat modeling
• Network segmentation and perimeter controls
• Identity and access management
• Encryption and data protection
• Security testing frameworks
• Security in deployment and operations

You build A security architecture with a threat model and controls mapped to each threat.

Generative AI can draft requirements, diagrams and reviews faster than people can. You use it for analysis work, check its output against the facts, and decide what an architect must never delegate.

• Drafting requirements and user stories with AI
• Generating and checking diagrams
• Using AI to review a design
• Verifying AI output against sources
• Confidentiality when sharing designs with a model
• What stays a human decision

You build A short analysis done with AI assistance, with each AI contribution verified.

Design the replacement for an aging system from the business need to the migration. You cover requirements, process change, architecture, data, security and a plan to get there, and present it to stakeholders.

• Understanding the legacy system and its users
• Setting requirements and quality targets
• Designing the new architecture and data flows
• Planning integration, migration and automation
• Presenting the design to stakeholders

You build A full replacement design for a legacy system, with a migration plan and a stakeholder presentation.