AWS SysOps Administrator -- operating workloads in production

An AWS SysOps professional keeps workloads on AWS available, performant, secure, recoverable and cost-effective after they have been deployed. Where a Solutions Architect decides how a system should be built, you answer whether it is working and how to keep it working: you carry the pager, investigate what alarms, run the changes and the patches, prove the backups restore, watch the capacity and the bill, and turn every repeated chore into automation so the next one costs less.

About 166 hours · 39 modules · 119 labs · certificate on completion · Cloud (AWS) and IT operations

For: administrator, engineer

Skills: aws, operations, monitoring, backup, incident-response

What the course covers

  1. Tier 1: how this tier works — Tier 1 turns you from someone who has clicked around the AWS console into someone who can stand up, connect, protect and recover a small estate from the command line, and explain what they did. It covers four modules: the account baseline (M1), compute and Session Manager (M2), networking (M3), and storage and databases (M4). It ends with a timed practical. Read this first: every lab creates billable resources in your own AWS account. Prices change and depend on region and usage, so this guide gives only unmeasured estimates in each cost note. Your own Billing console is the authority. Do the teardown at the end of each lab, every time.
  2. M1. The Account Baseline — Explain the shared responsibility model and say which side owns what for EC2, S3 and RDS. Lock down the root user: MFA on, no access keys, and used only for the few tasks that need it. Set up everyday administrator access through IAM Identity Center and sign in from the CLI with a short-lived SSO session. Create a cost budget with email alerts, and explain why a budget alerts but does not stop… (2 labs)
  3. M2. Compute with EC2 and Session Manager — Launch an EC2 instance from a template, with an instance profile, an encrypted gp3 root volume and IMDSv2 required. Explain EC2 instance states and which of them stop compute billing. Connect to an instance with Session Manager, with no inbound security-group rules and no SSH key. Resize an EBS volume while the instance runs, then grow the partition and file system. Take, tag and restore from an… (2 labs)
  4. M3. Networking with VPC — Plan a VPC address range and carve it into public and private subnets across two Availability Zones. Explain how route tables, internet gateways, NAT gateways and VPC endpoints decide where traffic goes. Contrast security groups (stateful, allow-only, on the interface) with network ACLs (stateless, ordered allow and deny, on the subnet). Deploy a two-tier VPC from a CloudFormation template and… (2 labs)
  5. M4. Storage and Databases (S3 and RDS) — Choose between block, object and managed database storage for a given workload. Create a private S3 bucket with Block Public Access, default encryption, versioning and lifecycle rules. Explain what a delete marker is, and recover both a deleted object and an overwritten one. Explain how lifecycle rules can make a recovery impossible, and read a lifecycle configuration. Launch a private RDS… (2 labs)
  6. Tier 1 exit practical: Hollowell Retail promo environment — Timed: 3 hours. Open book (your own notes and AWS documentation). A reviewer is needed for the fault step. (2 labs)
  7. Tier 2: how this tier works — Tier 2 is 48 hours across four modules: monitoring and remediation (M5), deployment and automation (M6), reliability and backup (M7), and security operations (M8). It is the heart of day-to-day cloud operations and covers most of what the AWS Certified CloudOps Engineer Associate exam (SOA-C03, formerly SysOps Administrator) tests. What you need before you start: you finished Tier 1, or you can build a two-tier VPC with EC2, S3 and RDS from the CLI. You have your own sandbox AWS account with a budget alarm already set. All labs default to the ca-central-1 region. Tools: AWS CLI v2 configured with an SSO or role profile, Python 3.12 with boto3, jq, Git, and cfn-lint (pip install cfn-lint). A terminal and a text editor are enough. Fake values: every account ID, access key and token in this material is obviously fake (for example 111122223333 and AKIAEXAMPLEFAKE). Never paste a real key into a lab file or a repository.
  8. M5. Monitoring, logging and remediation — Explain how CloudWatch organises metrics (namespace, metric name, dimensions, statistic, period) and read a metric correctly. Install and configure the CloudWatch agent to publish memory, disk and log data from an EC2 instance. Write alarms with the right evaluation window and missing-data behaviour, and combine them with a composite alarm. Query logs with CloudWatch Logs Insights and trace who… (2 labs)
  9. M6. Deployment, provisioning and automation — Describe a CloudFormation template (parameters, resources, outputs) and deploy it as a stack. Preview changes with a change set, detect drift, and recover from failed, stuck and rolled-back stacks. Lint a template before deployment and fix what the linter finds. Manage a fleet with Systems Manager: Run Command, Parameter Store, Patch Manager and maintenance windows. Write a small Python script… (2 labs)
  10. M7. Reliability and business continuity — Choose a load balancer type and configure target groups, health checks and deregistration delay. Build an Auto Scaling group that replaces failed instances and spreads across Availability Zones. Explain Route 53 routing policies and health checks, and RDS Multi-AZ, replicas and point-in-time recovery. Plan, run and restore from AWS Backup, S3 versioning and snapshots. Define RPO and RTO, choose… (2 labs)
  11. M8. Security and compliance operations — Explain how IAM evaluates a request, including trust policies, permission boundaries and explicit deny. Diagnose an access-denied error to the layer that caused it. Describe how KMS keys, key policies and envelope encryption work, and fix a key policy safely. Use detective services (GuardDuty, Inspector, Config, Security Hub, Access Analyzer) and say what each can and cannot see. Audit security… (2 labs)
  12. Practice exam (65 questions, 130 minutes) — This exam follows the style of the AWS Certified CloudOps Engineer Associate (SOA-C03) multiple-choice questions. Choose the one best answer. Aim for about two minutes per question, flag the hard ones and return to them. The answers are in the instructor guide, so mark your paper and then ask your instructor to score it.
  13. Tier 2 exit practical — This is a timed, individual assessment. It tests whether you can take over an unfamiliar, faulty environment and leave it monitored, recoverable and secure. There are no step-by-step instructions. You choose the method and you prove the result. (2 labs)
  14. Tier 3: how this tier works — Tier 3 is where you stop operating single servers and start operating systems: several networks joined together, containers and queues, a bill someone will ask you to explain, and a day when five things break at once. It covers four modules (M9 Advanced Networking, M10 Containers and Serverless Operations, M11 Cost and Performance Optimisation, M12 Incident Response and Operational Excellence) and ends with a case exam. Plan on about 44 hours in total, which is roughly 10 to 11 hours per module including the lab.
  15. M9. Advanced Networking — Explain when to use VPC peering, Transit Gateway and a VPN, and choose between them for a given set of networks. Read and reason about Transit Gateway route tables: association, propagation, static routes, blackholes and longest-prefix match. Build a segmented hub-and-spoke network where shared services can reach every spoke but the spokes cannot reach each other. Connect a simulated branch… (2 labs)
  16. M10. Containers and Serverless Operations — Explain how an ECS service, task definition, task and Fargate fit together, and the difference between the task execution role and the task role. Configure a rolling deployment with minimum healthy percent, maximum percent and the deployment circuit breaker with rollback, and explain what the breaker counts. Set up target-tracking autoscaling for a service and reason about cooldowns and capacity… (2 labs)
  17. M11. Cost and Performance Optimisation — Read a cost and usage extract, group it by service, resource and tag, and say where the money goes. Classify waste into categories (idle, oversized, orphaned, wrong type, unmanaged growth) and estimate savings without double counting. Distinguish waste from missing tags and from spend you cannot cut. Explain how Savings Plans work, size a commitment from the steady baseline, and calculate the… (2 labs)
  18. M12. Incident Response and Operational Excellence — Run the first ten minutes of an incident: declare, assign roles, assess severity, communicate and choose a first mitigation. Separate mitigation (stop the harm) from root cause (understand and fix it), and know which comes first. Use CloudTrail, CloudWatch, Auto Scaling activity, RDS metrics and GuardDuty findings as evidence, and build an accurate incident timeline from them. Contain a… (2 labs)
  19. Tier 3 exit case exam: Quillstack's three-account clean-up — Time-box: 4 hours, open book (your own notes and AWS documentation are allowed; no other person). Weight: this exam is 100 points and is required to complete Tier 3. No AWS account is needed: all data is inline and fictional. The account IDs, addresses and names are made up (addresses outside the 10.0.0.0/8 space come from ranges reserved for documentation). Region: ca-central-1. The prices in this exam are fictional exam rates, not AWS prices: use only the rates given here. (2 labs)
  20. Tier 4: how this tier works — Tier 4 is where you stop following labs and start running things. Four modules build the professional skills (about 38 hours in the course outline: M13 10, M14 10, M15 10, M16 8), and then the capstone has you take over a faulty production environment for a simulated 30 days. Everything you need for the capstone is printed inside this document. You do not need any other file.
  21. M13. Multi-account governance — Explain why a company with more than one team or environment should use more than one AWS account, and name the costs of doing so. Design an organizational unit (OU) structure for a small retailer and justify each account. Write, read and test service control policies (SCPs) and predict the effective permissions of a principal from the SCPs and IAM policies that apply to it. Set up human access… (2 labs)
  22. M14. CI/CD and infrastructure as code at scale — Describe a safe delivery pipeline for infrastructure from pull request to production, and say what each stage protects against. Give GitHub Actions access to AWS without storing long-lived keys, using OpenID Connect (OIDC) and a role with a tight trust policy. Write a least-privilege deployment arrangement: a limited deploy role plus a separate CloudFormation service role. Use cfn-lint and a… (2 labs)
  23. M15. Observability, SRE practice and on-call — Tell the difference between monitoring (is it broken?) and observability (why is it broken?), and name the signals you need for each. Choose a service level indicator (SLI), set a service level objective (SLO), and calculate the error budget. Build burn-rate alerts that page on fast budget loss and open tickets on slow loss. Cut a noisy alert set to a small, useful one without losing coverage.… (2 labs)
  24. M16. Exam prep, career and professionalism — Describe the structure of the AWS Certified CloudOps Engineer, Associate (SOA-C03) exam and plan study against its five domains. Read a scenario question, find the real requirement, and eliminate wrong answers by reasoning. Sit two timed practice exams, mark them, and turn the results into a gap report with a study plan. Answer operations interview questions with a clear structure, using real… (1 lab)
  25. Capstone: Bramblewood Outfitters — This section is everything you receive on Day 1. All data is fictional. Names, addresses, IP addresses (several come from ranges reserved for documentation), account numbers and key IDs are invented. Do not search for them and do not use real credentials with any of it. (2 labs)
  26. Stage 1. AWS and operations foundations — The role and its measures, AWS's infrastructure, operating-system administration, and operational access and records. (8 labs)
  27. Stage 2. Compute, storage and networking operations — EC2, Auto Scaling and load balancers, EBS, S3, EFS and FSx, VPCs, DNS and hybrid connectivity. (10 labs)
  28. Stage 3. Monitoring, logging and observability — CloudWatch, CloudTrail and AWS Config, alarms and dashboards, and application tracing. (6 labs)
  29. Stage 4. Identity and security operations — IAM administration and access troubleshooting, security findings, patching and vulnerability management. (6 labs)
  30. Stage 5. Automation and configuration management — Systems Manager, the CLI and scripting, infrastructure as code and change management. (6 labs)
  31. Stage 6. Availability, backup and disaster recovery — Operating for high availability, fault injection, AWS Backup, recovery and drills. (4 labs)
  32. Stage 7. Troubleshooting and incident management — A structured troubleshooting method, root-cause analysis, and running production incidents. (4 labs)
  33. Stage 8. Cost, performance and capacity management — Cost operations and FinOps, performance monitoring and capacity planning. (4 labs)
  34. Stage 9. Enterprise and advanced operations — Multi-account and hybrid operations, quotas, AWS Health, lifecycle and readiness. (4 labs)
  35. Stage 10. Practical mastery — Runbook engineering and operational maturity, then five realistic operational scenarios. (2 labs)
  36. Stage 11. Production incident simulations — Ten production incidents worked end to end: investigate from the evidence, restore service safely, validate recovery and write the incident report. (20 labs)
  37. Stage 12. Capstone — Operate an enterprise AWS environment through a week of incidents, changes and a recovery exercise. (1 lab)
  38. Projects (4 labs)
  39. Interview preparation (1 lab)

USD 449.25 · the first modules are free to try with an account.

Try it free