Task Statement 4.2: Design cost-optimized compute solutions.
📘AWS Certified Solutions Architect – (SAA-C03)
Hybrid compute means running compute workloads in both the cloud and on-premises environments. AWS provides solutions for customers who need cloud features but also want or need to keep some workloads close to their own data centers or on the edge. This is important for latency-sensitive applications, regulatory requirements, or intermittent connectivity.
The two main AWS hybrid compute options you need to know for the exam are:
1. AWS Outposts
What it is:
- AWS Outposts is basically an AWS rack of servers that you install in your own data center.
- It brings native AWS services, APIs, and tools directly to your on-premises location.
- You can run EC2 instances, EBS volumes, RDS databases, and other AWS services locally on your Outposts.
Key Features:
- Consistent AWS experience:
You use the same AWS APIs, tools, and management consoles as in the AWS cloud. This makes hybrid operations easier. - Low latency:
Because the servers are physically near your users or systems, applications that need real-time processing benefit from faster response times. - Integrated with AWS Cloud:
Data can move seamlessly to and from the AWS cloud when needed. - Managed by AWS:
AWS installs, monitors, and updates the hardware remotely, reducing operational complexity.
When to use it (exam perspective):
- Applications requiring low latency to on-premises systems.
- Data residency or compliance requirements that cannot move all data to the cloud.
- Hybrid apps where some workloads stay on-premises but still need AWS services.
Exam Tip: Know that Outposts is fully managed by AWS but runs on-premises, unlike a traditional server that you manage yourself.
2. AWS Snowball Edge
What it is:
- AWS Snowball Edge is a portable, rugged storage and compute device.
- It is designed for edge computing or offline environments where connectivity to AWS cloud may be limited or intermittent.
- You can run EC2 instances on the device itself, process data locally, and later transfer results to AWS cloud.
Key Features:
- Portable & rugged:
Snowball Edge can be shipped to remote locations, harsh environments, or disconnected sites. - Local processing:
You can run compute tasks on the device itself—for example, processing data from IoT devices or remote servers before sending it to the cloud. - Integrated with AWS services:
Data can later be imported into S3, Glacier, or other AWS services seamlessly. - Multiple device types:
There’s a Storage Optimized device for large datasets and a Compute Optimized device for heavier processing tasks.
When to use it (exam perspective):
- Locations with limited or no internet connectivity.
- Need local data processing at the edge.
- Large-scale data migration from remote sites to AWS.
Exam Tip: Remember Snowball Edge is both storage and compute at the edge, unlike Outposts, which is like a mini AWS data center on-premises.
Hybrid Compute Comparison Table (for Exam)
| Feature | AWS Outposts | AWS Snowball Edge |
|---|---|---|
| Location | Permanently installed on-premises | Portable, temporary/remote locations |
| Use Case | Hybrid cloud apps, low latency, compliance | Edge computing, offline processing, data migration |
| Managed by AWS | Yes | Yes (device setup & support) |
| Compute | EC2, RDS, Lambda (limited) | EC2 instances (compute workloads) |
| Connectivity | Continuous cloud connectivity | Can work offline; data uploaded later |
| Data Movement | Seamless with AWS Cloud | Batch upload or sync to AWS Cloud |
Key Exam Points to Remember
- Hybrid compute = cloud + on-premises.
- Outposts: permanent AWS infrastructure on-premises for low-latency, hybrid apps.
- Snowball Edge: portable compute + storage for edge locations, offline or disconnected processing.
- AWS manages the hardware in both cases.
- Use Outposts for always-on hybrid workloads, Snowball Edge for temporary, edge, or offline tasks.
✅ Memory Aid for Exam:
- Outposts = “Always On, On-Prem AWS”
- Snowball Edge = “Portable Cloud-in-a-Box”
