AWS Certified Solutions Architect – Associate Compute Services Syllabus

Compute Services syllabus — full chapter and topic list

Expand any chapter to see its topics and sub-topics. This is the whole examinable outline for Compute Services in AWS Certified Solutions Architect – Associate, not a summary of it.

1. 1

   Amazon EC2 Fundamentals

   5 topics

   • Instance families and selection
     • General purpose, compute, memory, storage and accelerated computing
   • AMIs and instance lifecycle
     • Custom AMIs, golden images and user data
   • EC2 purchasing options
     • On-Demand, Reserved Instances and Savings Plans
     • Spot Instances and Spot Fleets
     • Dedicated Hosts and Dedicated Instances
   • Placement groups (cluster, spread, partition)
   • EC2 networking (ENI, ENA, instance metadata service)
2. 2

   Elasticity & High Availability for Compute

   3 topics

   • Auto Scaling Groups
     • Launch templates and launch configurations
     • Dynamic, predictive and scheduled scaling policies
     • Health checks and instance refresh
   • Elastic Load Balancing
     • Application, Network and Gateway Load Balancers
     • Target groups, listeners and routing rules
     • Cross-zone load balancing and sticky sessions
   • Designing fault-tolerant, self-healing tiers
3. 3

   Containers on AWS

   4 topics

   • Amazon ECS
     • Tasks, services and task definitions
     • EC2 vs Fargate launch types
   • Amazon EKS for managed Kubernetes
   • Amazon ECR for image storage
   • AWS Fargate serverless containers
4. 4

   Serverless & Edge Compute

   4 topics

   • AWS Lambda
     • Event sources, triggers and execution model
     • Memory, timeout, concurrency and cold starts
     • Lambda layers and environment variables
   • Lambda integration patterns with other services
   • Lambda@Edge and CloudFront Functions
   • AWS Batch and Elastic Beanstalk overview

Compute Services flashcards for AWS Certified Solutions Architect – Associate

23 of 51 cards from the Compute Services deck — real questions with worked answers.

1. What are the five main EC2 instance family categories and their primary use cases?

   General Purpose (balanced, e.g. M/T) for web servers; Compute Optimized (C) for high-CPU/batch; Memory Optimized (R/X) for in-memory databases; Storage Optimized (I/D/H) for high disk throughput/IOPS; Accelerated Computing (P/G/Inf) for GPU/FPGA/ML workloads.

2. What is the difference between burstable T-family instances and other general-purpose instances?

   T-family (e.g. T3) accumulate CPU credits during idle periods and spend them to burst above a baseline. Standard mode caps usage at credits available; Unlimited mode allows bursting beyond credits for an extra charge. Other families provide fixed, consistent CPU performance.

3. What is an Amazon Machine Image (AMI) and what does it contain?

   An AMI is a template used to launch EC2 instances. It contains the root volume snapshot (OS, application server, applications), launch permissions controlling which accounts can use it, and a block device mapping specifying volumes attached at launch.

4. What are the four AMI categories by source/ownership?

   AWS-provided (e.g. Amazon Linux), AWS Marketplace AMIs (vendor, may carry software charges), Community AMIs (publicly shared by users), and your own custom/private AMIs.

5. Are AMIs region-specific, and how do you use one in another region?

   Yes, AMIs are scoped to a single region. To use one in another region you must copy the AMI to the target region, which creates a new AMI with a new ID there.

6. List the main EC2 instance lifecycle states.

   pending (launching), running, stopping, stopped, rebooting, shutting-down, and terminated. Stopped instances can be restarted; terminated instances are permanently deleted.

7. What happens to instance store data when an EC2 instance is stopped versus terminated?

   Instance store (ephemeral) data is lost on stop AND terminate. EBS root volumes persist on stop and can be deleted or retained on terminate based on the DeleteOnTermination flag.

8. What changes about an EC2 instance's public IP when it is stopped and started?

   A default auto-assigned public IPv4 address is released on stop and a new one is assigned on start. To keep a fixed public IP, attach an Elastic IP, which persists across stop/start.

9. What are the main EC2 purchasing (pricing) options?

   On-Demand (pay per second/hour, no commitment), Reserved Instances (1 or 3 year commitment for discount), Savings Plans (commit to $/hour spend), Spot Instances (spare capacity at deep discount, interruptible), and Dedicated Hosts/Instances (physical isolation).

10. What discount and commitment characterize Reserved Instances, and what are the offering classes?

    RIs offer up to ~72% off On-Demand for a 1- or 3-year term. Standard RIs give the biggest discount but limited changes; Convertible RIs allow changing instance attributes/family for a smaller discount. Payment can be No, Partial, or All Upfront.

11. How much can Spot Instances save and what is the key risk?

    Spot can save up to ~90% off On-Demand using spare capacity, but AWS can reclaim instances with a 2-minute interruption notice when capacity is needed or price exceeds your max. Best for fault-tolerant, stateless, flexible workloads.

12. Compare Savings Plans vs Reserved Instances.

    Both give discounts for 1- or 3-year commitments. RIs commit to specific instance attributes; Savings Plans commit to a consistent $/hour spend and are more flexible. Compute Savings Plans apply across instance family, region, OS, tenancy, and even Fargate/Lambda; EC2 Instance Savings Plans are limited to a family in a region for a larger discount.

13. What is a Spot Fleet and a Spot Instance request type?

    A Spot Fleet is a collection of Spot (and optionally On-Demand) instances meeting a target capacity, optimized by allocation strategy (e.g. lowest-price, capacity-optimized). Spot requests can be one-time or persistent (re-requested after interruption).

14. What is the purpose of EC2 placement groups?

    Placement groups control how instances are physically placed on underlying hardware to optimize for low latency/high throughput (cluster), high availability (spread), or large distributed workloads with fault isolation (partition).

15. Describe a cluster placement group and its trade-off.

    A cluster placement group packs instances close together in a single Availability Zone for the lowest network latency and highest throughput (up to 10/100 Gbps). Trade-off: if the rack fails, all instances may fail together, so it sacrifices availability for performance.

16. Describe a spread placement group and its key limit.

    A spread placement group places each instance on distinct underlying hardware (separate racks with independent power/network) to minimize correlated failures. Limited to 7 running instances per Availability Zone per group. Best for small numbers of critical instances.

17. Describe a partition placement group.

    A partition placement group divides instances into logical partitions, each on a distinct rack with isolated power/network. Up to 7 partitions per AZ, many instances per partition. Ideal for large distributed/replicated workloads like HDFS, Cassandra, and Kafka where partition-level fault isolation matters.

18. What is an Elastic Network Interface (ENI)?

    An ENI is a virtual network card in a VPC that has a primary private IPv4, optional secondary IPs, an Elastic IP, a MAC address, and security groups. It can be attached/detached and moved between instances in the same AZ, useful for failover.

19. What is Elastic Network Adapter (ENA) and what does it enable?

    ENA is the technology behind Enhanced Networking, providing higher bandwidth, higher packet-per-second performance, and lower/consistent latency via SR-IOV. Supported instances can reach up to 100 Gbps with ENA.

20. What is the Instance Metadata Service (IMDS) and its endpoint?

    IMDS lets an instance query its own metadata (instance ID, IAM role credentials, user data, networking info) via the link-local address http://169.254.169.254/latest/meta-data/. It is reachable only from within the instance.

21. What is the difference between IMDSv1 and IMDSv2?

    IMDSv1 is a request/response model that is vulnerable to SSRF attacks. IMDSv2 is session-oriented and token-based: you first PUT to obtain a token, then send it in subsequent requests, mitigating SSRF. AWS recommends enforcing IMDSv2.

22. What is an EC2 Auto Scaling Group (ASG)?

    An ASG automatically launches and terminates EC2 instances to maintain a desired capacity across one or more AZs, replacing unhealthy instances and scaling in/out based on demand. It uses a Launch Template (or Launch Configuration) to define instances.

23. What are the three capacity settings of an Auto Scaling Group?

    Minimum (never go below), Desired (target number to maintain), and Maximum (never exceed). The ASG keeps the running count at desired, bounded by min and max.

See more Compute Services flashcards →

Planning Compute Services for AWS Certified Solutions Architect – Associate

Compute Services (AWS Certified Solutions Architect – Associate) FAQ