Data Migration with Snow Family and DataSync

Design a data migration strategy using AWS Snow Family and DataSync.

Introduction

30 Second Summary

Imagine your company has 100 TB of critical data sitting in an on-premises data center. You need it all in the cloud, but your internet connection would take over 3 months to transfer it. Choosing the wrong migration method means blown deadlines and wasted budgets.

In this project, you will build a working AWS DataSync transfer between S3 buckets and develop a decision framework for choosing between AWS Snow Family devices based on real-world constraints like data volume, bandwidth, and compute needs.

What You'll Build

You will run a live DataSync task that moves files across AWS regions, then work through realistic migration scenarios to match each one to the correct transfer service.

By the end of this project, you'll have:

  • A working DataSync task that transfers data between S3 buckets across AWS regions, showing you the full task lifecycle from preparation through verification.
  • A decision framework for choosing between Snowcone, Snowball Edge, Snowmobile, and DataSync based on data volume, bandwidth, timeline, and compute requirements.
  • The ability to calculate transfer times and identify the breakpoint where online transfer becomes impractical and offline shipping takes over.
  • Secret Mission: Design a hybrid edge-to-cloud pipeline that combines Snow Family with DataSync for a manufacturing IoT scenario.

Are there any prerequisites?

You need an AWS account and basic familiarity with S3 and the AWS Console. All work in this project happens directly in the AWS Management Console.

Before We Start

Before we dive in, let's lock in what you're about to build and why it matters. Choosing the wrong data transfer method can mean weeks of wasted time or thousands in unnecessary costs.

Create Your S3 Buckets for DataSync

To perform a real data transfer with DataSync, you need storage on both ends of the pipeline. A source bucket holds the data you want to move. A destination bucket receives it.

Creating these buckets in different AWS regions simulates a cross-region disaster recovery pattern. This is one of the most common real-world uses of DataSync, and exactly the kind of scenario you will face on the SAA-C03 exam.

In this step, get ready to:
  • Create a source S3 bucket in us-east-1 and a destination bucket in us-west-2.
  • Upload test files to the source bucket.
  • Frame the media company migration scenario that drives the rest of this project.
Create your source and destination buckets

You need two buckets in different regions so DataSync can transfer data between them. This mirrors a real disaster recovery setup where data in one region is replicated to another for safety.

  • In the AWS Console, type S3 in the top search bar and select S3 from the results.
  • In the left navigation pane, choose General purpose buckets.
  • Click Create bucket.
  • For Bucket name, enter a unique name starting with datasync-source- followed by your initials and today's date (for example, datasync-source-jm-20260724).
  • For AWS Region, select US East (N. Virginia) us-east-1.
  • Leave all other settings at their defaults.
  • Click Create bucket at the bottom of the page.

You should see a green success banner and your new bucket listed on the General purpose buckets page.

  • Click Create bucket again to create the destination bucket.
  • For Bucket name, enter a name starting with datasync-dest- followed by the same initials and date (for example, datasync-dest-jm-20260724).
  • For AWS Region, select US West (Oregon) us-west-2.
  • Leave all other settings at their defaults.
  • Click Create bucket.

You should now see both buckets listed on the General purpose buckets page.

Why different regions?

Placing the source bucket in us-east-1 and the destination in us-west-2 simulates a cross-region disaster recovery pattern. If the primary region goes down, data is already available in the backup region.

DataSync handles all the complexity of cross-region transfers for you, including encryption in transit and data integrity verification.

Seeing a bucket name error?

S3 bucket names must be globally unique across all AWS accounts. If you see an error saying the name is already taken, try adding extra characters to your bucket name (for example, add a random number at the end).

Bucket names must be lowercase, between 3 and 63 characters, and can only contain lowercase letters, numbers, and hyphens.

Help me fix my S3 bucket naming error.

Upload test files to the source bucket

DataSync needs actual data to transfer. You will upload a few small test files to the source bucket so you have something real to move in the next step.

  • On the General purpose buckets page, click the name of your source bucket (the one in us-east-1) to open it.
  • On the Objects tab, click Upload.
  • Click Add files and select 3 to 5 small files from your computer (text files, images, or PDFs all work).
  • Click Upload at the bottom of the page.

You should see a green success banner confirming your files uploaded successfully.

  • Click Close to return to your bucket.

Before you check your work, predict what you will see. How many objects should be listed on the Objects tab of your source bucket?

You should see all of your uploaded files listed in the Objects tab, matching the number of files you selected.

Don't see your uploaded files?

Make sure you are looking at the source bucket, not the destination bucket. Check the region indicator in the top navigation bar to confirm you are viewing the bucket in us-east-1.

If the upload failed, try again with smaller files. Each file should be under 5 MB for this exercise.

Help me troubleshoot my S3 file upload.

Understand the migration scenario

Now that your storage is ready, let's frame the business scenario that drives this project.

What is the scenario?

You are a solutions architect for a media company. They have 500 GB of video assets in one AWS region that need to be replicated to another region for disaster recovery. The data changes daily.

This is an online transfer scenario. The company has good network connectivity between regions, the data changes frequently, and they need ongoing replication. DataSync is the right choice here because it handles incremental, scheduled transfers over the network.

Your test buckets simulate this pattern on a small scale. The source bucket (us-east-1) represents the primary region. The destination bucket (us-west-2) represents the disaster recovery region.

Why DataSync instead of S3 Cross-Region Replication?

DataSync gives you full control over when and how transfers run. You can schedule tasks, filter which files transfer, verify data integrity, and monitor progress in detail.

S3 Cross-Region Replication is simpler but automatic. For the SAA-C03 exam, you need to know when each tool is the better fit. DataSync excels when you need task-based, on-demand, or scheduled transfers with detailed monitoring.

Your storage is set up and the scenario is clear. Next up, you will create a DataSync task to actually move data between these buckets and watch the entire transfer lifecycle in action.

Transfer Data with AWS DataSync

Your source and destination buckets are ready. Now you need to actually move data between them.

AWS DataSync is the service designed for exactly this. It handles online data transfers between AWS storage services, managing the full lifecycle of preparing, transferring, and verifying your data. In this step, you will create a real DataSync transfer between your two S3 buckets.

In this step, get ready to:
  • Create DataSync locations for your source and destination S3 buckets.
  • Configure and run a DataSync task to transfer your files across regions.
  • Verify the transfer completed successfully with a status of Success.
Create your DataSync source location

DataSync uses the concept of locations to define where data comes from and where it goes. You need to create a location for each bucket before you can transfer data between them.

  • Open the DataSync Console by navigating to https://console.aws.amazon.com/datasync/ in your browser.
  • In the top navigation bar, switch your region to US West (Oregon) us-west-2. This is the region of your destination bucket.
  • In the left navigation pane, expand Data transfer, then choose Locations.
  • Click Create location.
  • For Location type, choose Amazon S3, and then choose General purpose bucket.
  • For S3 URI, select your source bucket (the one in us-east-1 with your test files).
  • For IAM role, choose Autogenerate. DataSync will automatically create an IAM role with the permissions required to access the S3 bucket.
  • Click Create location.

You should see a confirmation that your source location was created successfully. It now appears in the Locations list.

What is a DataSync location?

A location describes where you are copying data from or to. Every DataSync transfer task needs exactly two locations: a source and a destination. DataSync supports S3 buckets, EFS file systems, FSx file systems, and on-premises storage as locations.

Not seeing your source bucket in the list?

Your source bucket is in us-east-1, but you should still see it in the bucket dropdown. Make sure you selected General purpose bucket (not Directory bucket). If the bucket still does not appear, double-check that you are signed into the correct AWS account.

Need more help? Why can't I see my S3 bucket when creating a DataSync location?

Create your DataSync destination location

Now create a second location for your destination bucket. The process is identical.

  • In the left navigation pane under Data transfer, choose Locations, then click Create location.
  • For Location type, choose Amazon S3, then General purpose bucket.
  • For S3 URI, select your destination bucket (the one in us-west-2).
  • For IAM role, choose Autogenerate.
  • Click Create location.

You should now see two locations in your Locations list. One points to your source bucket in us-east-1, and the other to your destination bucket in us-west-2.

Seeing an error when creating the destination location?

Make sure you selected a different bucket from the one you used for the source location. Each location must point to a distinct S3 URI. If you see a permissions error, confirm that the IAM role is set to Autogenerate.

Still stuck? Help me troubleshoot my DataSync destination location creation error.

Create and run your DataSync task

With both locations configured, you can now create a task that connects them and moves your data.

  • In the left navigation pane, expand Data transfer, then choose Tasks.
  • Click Create task.
  • On the Configure source location page, select Choose an existing location.
  • Select your source location (the us-east-1 bucket), then click Next.
  • On the Configure destination location page, select Choose an existing location.
  • Select your destination location (the us-west-2 bucket), then click Next.
  • On the Configure settings page, enter cross-region-dr-sync as the task name.
  • For Task mode, select Enhanced.

Why Enhanced mode?

Enhanced mode parallelizes the data transfer process: listing, preparing, transferring, and verifying all happen simultaneously. For S3-to-S3 transfers, Enhanced mode supports virtually unlimited object counts and delivers higher performance than Basic mode.

  • Leave all other settings as defaults and click Next.
  • On the Review page, review your settings and click Create task.

Your task is now created. You should land on the task details page with a status of Available.

  • On the task details page, click Start.
  • Choose Start with defaults.

Before you watch the task run, think about this: what stages do you expect the task to go through before it reports success?

Watch the task execution progress through these statuses: Launching, Preparing, Transferring, Verifying, and finally Success. With your small test files, this should complete in under a minute.

What do these statuses mean?

Each status represents a stage in the transfer lifecycle. Launching initializes the task resources. Preparing scans the source to determine what needs transferring. Transferring moves the actual data. Verifying compares checksums between source and destination to confirm data integrity.

Enhanced mode runs these stages in parallel for large datasets, but with your small test files they appear sequential.

  • Once the task shows Success, navigate to the S3 Console.
  • Switch to the us-west-2 region and open your destination bucket.
  • Confirm that your destination bucket now contains the same files you uploaded to the source bucket.

Your destination bucket should now have all the test files that were in your source bucket. DataSync has successfully transferred and verified your data across regions.

Task showing an error status?

If your task shows an Error status instead of Success, check the execution details for an error message. The most common cause is an IAM permissions issue. Confirm that both locations were created with Autogenerate for the IAM role. You can delete the task, recreate your locations with autogenerated roles, and try again.

Need help? My DataSync task is showing an error status. How do I troubleshoot it?

You have a working cross-region data transfer. Next up, you will push this approach to its limits and discover what happens when online transfers are not fast enough.

Hit the Bandwidth Wall

Your DataSync task just transferred files between regions in seconds. For a handful of test files, online transfer is fast and painless.

But what happens when the dataset is not a few kilobytes, but 100 terabytes? In this step, you will calculate real transfer times and discover why online-only migration breaks down at scale. This is the moment where AWS Snow Family enters the picture.

In this step, get ready to:
  • Calculate transfer times for three real-world data migration scenarios.
  • Identify the breakpoint where online transfer becomes impractical.
  • Connect this decision to SAA-C03 exam domains.
Calculate transfer times for large datasets

The formula for estimating data transfer time is simple: divide total data size by available bandwidth. The result tells you how long a pure online transfer would take under ideal conditions.

Before you calculate, think about this: your media company has 100 TB of video assets and a 100 Mbps dedicated internet connection. Do you think DataSync could handle this in a reasonable timeframe?

  • Work through Scenario A using the calculation below:
Scenario A: 100 TB over a 100 Mbps connection

  Step 1: Convert to the same unit (gigabits)
  100 TB = 800,000 Gb

  Step 2: Convert bandwidth to Gbps
  100 Mbps = 0.1 Gbps

  Step 3: Divide data by bandwidth
  800,000 Gb / 0.1 Gbps = 8,000,000 seconds

  Step 4: Convert to days
  8,000,000 seconds / 86,400 seconds per day = ~92 days

92 days. That is over three months of continuous, dedicated transfer with zero interruptions. For a disaster recovery scenario, waiting three months is not an option.

  • Now work through Scenario B:
Scenario B: 50 TB over a 1 Gbps connection

  50 TB = 400,000 Gb
  400,000 Gb / 1 Gbps = 400,000 seconds
  400,000 seconds / 86,400 = ~4.6 days
  • Finally, work through Scenario C:
Scenario C: 5 TB over a 100 Mbps connection

  5 TB = 40,000 Gb
  40,000 Gb / 0.1 Gbps = 400,000 seconds
  400,000 seconds / 86,400 = ~4.6 days

What do these numbers mean?

Scenario A (92 days) is clearly unacceptable for any business with a deadline. No company can wait three months for a migration.

Scenarios B and C (4.6 days each) are borderline. They might work if the bandwidth can be fully dedicated to the transfer with no other traffic competing for the line.

These calculations assume ideal conditions. In reality, network overhead, retries, and competing traffic make actual transfer times longer.

Identify the breakpoint

AWS recommends considering Snow Family when an online transfer would take more than approximately one week. This is the key decision threshold for the SAA-C03 exam.

  • Apply the one-week rule to your three scenarios:
The one-week breakpoint:

  Scenario A: 92 days   --> Far exceeds one week. Snow Family needed.
  Scenario B: 4.6 days  --> Under one week, but borderline.
                            Works if bandwidth is fully dedicated.
  Scenario C: 4.6 days  --> Under one week, but borderline.
                            Depends on whether bandwidth can be dedicated.

Why one week?

One week is the practical threshold where network-based transfer starts to become risky. Connections can drop, bandwidth can fluctuate, and the business cannot afford to restart a multi-day transfer from scratch.

Snow Family devices are physically shipped to you, loaded with data, and shipped back to AWS. For large datasets, this physical shipping approach is faster than waiting for a network transfer to complete.

  • Return to the DataSync Console and select your cross-region-dr-sync task.
  • Choose **History** to view the execution details from your earlier transfer.
  • Note how quickly your small test files transferred compared to the 92-day estimate for 100 TB.

What should I see?

Your task execution shows a status of Success with the total duration measured in seconds. The bytes transferred matches your small test files. This speed is possible because you had minimal data and plenty of bandwidth available.

Understand why this matters for the exam

The SAA-C03 exam tests this exact decision-making process across two domains.

  • **Domain 4, Task 4.1** explicitly tests determining the lowest cost method of transferring data for a workload to AWS storage.
  • **Domain 3, Task 3.5** tests data transfer services with appropriate use cases, including sizes and speeds needed to meet business requirements.

How does this appear on the exam?

Exam questions typically present a scenario with a specific data volume, available bandwidth, and a deadline. You calculate whether online transfer can meet the deadline. If it cannot, the answer involves Snow Family.

The question will never ask you to do the math on paper. Instead, you need to recognize the pattern: massive dataset plus limited bandwidth plus tight deadline equals Snow Family.

You have now identified the fundamental limitation of online data transfer. Your DataSync task worked perfectly for small files, but the math proves it breaks down at scale. Next up, you will learn which Snow Family device to choose for different scenarios.

Choose the Right Snow Device

In the previous step, you discovered that online transfers break down at scale. A 100 TB migration over 100 Mbps would take 92 days.

Now you need to know which Snow Family device to recommend. Each device has different capacity, compute power, and use case characteristics. The SAA-C03 exam tests these differences directly.

In this step, get ready to:
  • Map all four Snow Family devices to their correct use cases.
  • Match five practice scenarios to the correct device or DataSync.
  • Understand the security features shared across all Snow devices.
Map the Snow Family devices

AWS offers four physical devices for offline data transfer. Each one targets a different combination of storage capacity, compute power, and portability.

Think of it as a size ladder. The smallest device fits in a backpack. The largest requires a semi-trailer truck.

AWS Snowcone
  • Storage: 8 TB usable.
  • Compute: 2 vCPUs, 4 GB RAM.
  • Ultra-portable at 4.5 lbs (fits in a backpack).
  • Use for: edge IoT collection, tactical computing, transfers under 32 TB (order multiple devices).
  • Unique feature: can run DataSync on the device itself for online transfer back to AWS.
AWS Snowball Edge Storage Optimized
  • Storage: 80 TB HDD + 1 TB SATA SSD.
  • Compute: 40 vCPUs, 80 GiB RAM.
  • S3-compatible object storage on the device.
  • Use for: large-scale data migrations and capacity-oriented workloads where storage is the primary need.
AWS Snowball Edge Compute Optimized
  • Storage: 42 TB HDD + 7.68 TB NVMe SSD.
  • Compute: 52 vCPUs, 208 GiB RAM.
  • Optional NVIDIA Tesla V100 GPU for ML inference.
  • Use for: ML inference at the edge, video analytics, and edge computing scenarios where you need serious compute alongside moderate storage.
AWS Snowmobile
  • Storage: up to 100 PB per container.
  • A 45-foot ruggedized shipping container pulled by a semi-trailer truck.
  • AWS personnel handle all logistics. GPS tracking and 24/7 video surveillance.
  • Use for: exabyte-scale migrations (think entire data centers or media vaults).
  • Not available through self-service console. You must contact AWS Sales.

How do I remember which device is which?

Think of a size and purpose ladder. Snowcone is the backpack device for edge collection. Snowball Edge Storage Optimized is the suitcase-sized workhorse for bulk migrations. Snowball Edge Compute Optimized is the same form factor but trading storage for GPU power. Snowmobile is a literal truck for when nothing else scales.

On the exam, the keywords that point to each device are: "portable" or "IoT" (Snowcone), "large migration" or "80 TB" (Storage Optimized), "ML at edge" or "GPU" (Compute Optimized), and "exabyte" or "petabyte-scale" (Snowmobile).

Apply decision criteria to practice scenarios

The SAA-C03 exam presents migration scenarios and asks you to pick the right service. Let's practice with five scenarios that cover each device.

  • Read each scenario below and decide which Snow Family device (or DataSync) you would recommend before revealing the answer.

Scenario 1: A hospital needs to collect 6 TB of patient imaging data from a remote clinic with no internet connection.

Scenario 2: A financial firm is decommissioning a data center with 60 TB of archived records.

Scenario 3: An autonomous vehicle company needs to process 40 TB of lidar data at a remote testing site, running ML inference locally before shipping to AWS.

Scenario 4: A media company is migrating their entire 50 PB film vault to AWS.

Scenario 5: A company generates 2 TB of log data daily and has a 1 Gbps Direct Connect link.

Before you check the answers, write down your choice for each scenario. What device did you pick, and why?

✔️ I'll give it a go!

Write down your five answers, then switch to the other tab to check your work.

ⓧ Maybe a little help...

Here are the key decision factors to consider for each scenario:

  • How much data needs to move? (This narrows device options by capacity.)
  • Is there internet connectivity? (No internet means offline/physical only.)
  • Does data need local processing before transfer? (This points to Compute Optimized.)
  • Is the transfer ongoing or one-time? (Ongoing with good bandwidth = DataSync.)
Check your answers

Scenario 1 - Snowcone. The data is under 8 TB, the location is remote with no internet, and the device needs to be portable enough to carry to a clinic. All three keywords point to Snowcone.

Scenario 2 - Snowball Edge Storage Optimized. The primary goal is a large data migration (60 TB). No edge compute is needed. The Storage Optimized variant handles up to 80 TB and is built for exactly this use case.

Scenario 3 - Snowball Edge Compute Optimized. The company needs ML inference locally (GPU compute at the edge) plus moderate storage (40 TB fits within 42 TB). The Compute Optimized variant with its optional NVIDIA Tesla V100 GPU is the right fit.

Scenario 4 - Snowmobile. At 50 PB, this is exabyte-scale. No combination of Snowball Edge devices is practical at this volume. Snowmobile handles up to 100 PB per container.

Scenario 5 - DataSync. The transfer is ongoing (daily), incremental, and the company has a 1 Gbps Direct Connect link. Good bandwidth plus recurring transfers means DataSync is the right choice. No physical device needed.

Why does Scenario 4 use Snowmobile instead of many Snowball Edge devices?

At 50 PB, you would need over 625 Snowball Edge Storage Optimized devices (50,000 TB / 80 TB each). Managing that many shipments, loading cycles, and return logistics is impractical. Snowmobile handles up to 100 PB in a single container with AWS managing all logistics.

Why this matters for the exam

SAA-C03 Domain 4 (Task 4.1) explicitly tests "Determining the lowest cost method of transferring data for a workload to AWS storage." Domain 3 (Task 3.5) tests "Data transfer services with appropriate use cases" and "Sizes and speeds needed to meet business requirements."

On the exam, read the scenario carefully for data volume, connectivity, compute needs, and whether the transfer is one-time or ongoing. These four factors uniquely identify the correct service every time.

Understand security across all Snow devices

Security is another area the exam tests. All Snow Family devices share the same set of security features regardless of size.

  • All data is encrypted with AWS KMS keys. The encryption keys are never stored on the device itself.
  • Tamper-evident and tamper-resistant enclosures protect the device during shipping. A Trusted Platform Module (TPM) detects unauthorized access.
  • E Ink shipping labels provide chain-of-custody tracking. The label updates automatically so the device can only be shipped to the correct AWS facility.
  • After your data is transferred to your S3 bucket, AWS erases the device following NIST media sanitization standards.

Exam keyword: encryption keys

If a question asks "where are the encryption keys stored for Snow Family devices," the answer is always AWS KMS. The keys are never stored on the physical device. This is a key distinction that prevents data exposure if the device is intercepted during shipping.

Let's confirm you can match each scenario to the correct service. Look back at your written answers from the five scenarios above.

  • Verify that you matched all five correctly: Snowcone, Snowball Edge Storage Optimized, Snowball Edge Compute Optimized, Snowmobile, and DataSync.
  • For each answer, confirm you can explain the reasoning in one sentence (data volume, connectivity, compute needs, or transfer frequency).

Got one or more wrong?

The most common mix-up is between Snowball Edge Storage Optimized and Compute Optimized. Remember: if the scenario mentions ML, GPU, video analytics, or local processing, it is Compute Optimized. If it only mentions moving a large amount of data, it is Storage Optimized.

Another common mistake is choosing Snow Family when DataSync would work. If the scenario mentions ongoing/daily transfers with good bandwidth, DataSync is the answer.

Help me understand when to use Snowball Edge Storage Optimized vs Compute Optimized

You now have a complete decision framework for choosing between all four Snow Family devices and DataSync. Next up, you will explore the Snow Family Console to see how the ordering and job creation workflow looks in practice.

Explore Snow Family in the AWS Console

You matched real-world migration scenarios to the correct Snow Family device in the previous step. Now it's time to see how these devices are actually ordered and managed inside the AWS Console.

The Snow Family Console is where solutions architects create jobs to order physical devices. Walking through the Create Job wizard reveals exam-relevant details like job types and device options that appear directly on the SAA-C03.

In this step, get ready to:
  • Navigate the Snow Family Console and open the Create Job wizard.
  • Identify the three job types available for Snow Family devices.
  • Understand how DataSync and Snow Family combine for edge-to-cloud transfers.
Navigate to the Snow Family Console

The Snow Family Console is separate from the S3 or DataSync consoles you used earlier. It is the interface where AWS customers order physical Snow devices for offline data transfers and edge computing.

  • In the AWS Console search bar at the top, type Snow Family and select the Snow Family service from the results.
  • On the Snow Family home page, click Order an AWS Snow Family device (if this is your first time) or click Create Job in the left navigation bar.

Before you proceed into the wizard, think about this: how many distinct job types do you think AWS offers for Snow Family devices?

  • On the Plan your job page, observe the three job types listed:
  • Import into Amazon S3 - AWS ships an empty device to you. You load data onto it and ship it back. Your data uploads to S3.
  • Export from Amazon S3 - AWS loads your S3 data onto a device and ships it to you. You copy data to your local servers.
  • Local compute and storage only - Perform compute and storage workloads on the device without transferring data to or from AWS.

What do these job types mean for the exam?

Import jobs move data TO AWS. Export jobs move data FROM AWS. Local compute jobs keep everything on the device at the edge.

The SAA-C03 expects you to know which job type to recommend based on the scenario's data flow direction. A data center migration to AWS is always an Import job. Sending reference data to a remote site is an Export job.

Explore device selection

Now select a job type to see which physical devices AWS offers for that workflow.

  • Select Import into Amazon S3 as the job type.
  • Click Next to proceed to the device selection step.
  • Observe the device options presented on the page. You should see three choices:
  • Snowcone - 8 TB usable storage, ultra-portable at 4.5 lbs.
  • Snowball Edge Storage Optimized - 80 TB HDD capacity for large-scale migrations.
  • Snowball Edge Compute Optimized - 42 TB HDD with optional GPU for ML at the edge.

Where is Snowmobile?

Snowmobile does not appear in the self-service console. Because Snowmobile involves a 45-foot shipping container and AWS-managed logistics, you must contact AWS Sales directly to arrange one. On the exam, remember that Snowmobile is the answer only for exabyte-scale migrations (10+ PB).

  • Note the listed capacities and features for each device. Compare them to the decision framework you built in the previous step.
  • Cancel out of the wizard by navigating away or closing the page. Do NOT complete the job creation. No charges are incurred.

No charges from browsing

Exploring the Create Job wizard does not create any resources or cost anything. A job is only created (and charges begin) when you click Create job on the final review page and AWS begins preparing a device for shipment.

Connect the dots to DataSync

Snow Family and DataSync are not competing services. They complement each other and can even work together on the same device.

  • Online path (DataSync from Snowcone): Connect a Snowcone to your local network, run a DataSync agent on the device, and transfer data directly to S3, EFS, or FSx over the network.
  • Offline path (physical shipping): Load data onto any Snow device, ship it to AWS, and AWS uploads the data into your S3 bucket.

Why two paths from the same device?

Snowcone is unique in the Snow Family because it supports both online and offline transfer. A field researcher might collect 6 TB of sensor data at a remote site with limited bandwidth. They could use DataSync on the Snowcone to trickle summarized results to AWS in real time, then ship the full raw dataset offline when the device is full.

The other Snow devices (Snowball Edge Storage Optimized, Snowball Edge Compute Optimized) transfer data to AWS through physical shipping only.

This gives you the complete picture for the exam. You now have two decision axes: online vs offline transfer method, and which specific device or service fits the scenario's constraints.

You can now navigate to the Snow Family Console, identify the three job types, observe the available device options, and explain how DataSync bridges the gap between edge devices and the cloud. That is the full Snow Family picture the SAA-C03 expects you to know.

Secret mission

Design a Hybrid Edge-to-Cloud Pipeline

You have mastered DataSync for online transfers and Snow Family for offline bulk moves. Now combine both into a single architecture for a manufacturing company with 12 factory sites, limited bandwidth, and real-time edge processing needs.

Clean Up Your Resources

Clean Up Your Resources

Decide whether to keep your resources running, pause them to come back later, or delete them entirely. This project uses minimal S3 storage for test files and a DataSync task that incurs no cost when idle, so ongoing charges are negligible (a few cents per month at most).

Cost warning

Your two S3 buckets store only a handful of small test files. The storage cost is pennies per month. However, if you leave the buckets indefinitely, those small charges accumulate. Delete them if you do not plan to revisit this project.

Resources you used:

  • S3 source bucket (datasync-source-[initials]-[date]) in us-east-1.
  • S3 destination bucket (datasync-dest-[initials]-[date]) in us-west-2.
  • DataSync task (cross-region-dr-sync) and two S3 locations.
  • Two autogenerated IAM roles with a DataSync prefix.

Keep everything running

No action needed. Choose this if you plan to experiment further with DataSync or want to keep your test data for reference.

  • Your S3 storage costs will be a few cents per month for the small test files.
  • The DataSync task costs nothing when it is not running.
  • The IAM roles remain in your account but incur no charges.
  • The Snow Family Console exploration created no resources.

Pause - I'll come back to this later

Shut down running processes to free up memory and minimize costs while keeping your resources intact for later.

  • No running processes to stop. DataSync tasks only run on demand.
  • Your S3 buckets and DataSync resources remain available at negligible cost.
  • You can return and re-run your DataSync task or create new tasks at any time.

Delete - I don't want to use this again

Remove all project resources and start fresh. Follow these steps in order.

  • Delete the DataSync task: Open the DataSync Console at https://console.aws.amazon.com/datasync/. In the left navigation pane, expand Data transfer, then choose Tasks. Select your cross-region-dr-sync task. Choose Actions, then Delete. Confirm by choosing Delete in the dialog box.
  • Delete the DataSync locations: In the left navigation pane, expand Data transfer, then choose Locations. Select each location, choose Delete, confirm by entering delete, then choose Delete.
  • Empty and delete the S3 source bucket: Open the S3 Console. In the left navigation pane, choose General purpose buckets. Select your source bucket, then choose Empty. Type permanently delete to confirm, then choose Empty. After emptying, select the bucket again, choose Delete, enter the bucket name to confirm, and choose Delete bucket.
  • Empty and delete the S3 destination bucket: Repeat the same process for your destination bucket in us-west-2.
  • Delete the autogenerated IAM roles: Open the IAM Console at https://console.aws.amazon.com/iam/. In the navigation pane, choose Roles. Search for roles with DataSync in the name. Select each role, choose Delete at the top, enter the role name to confirm, and choose Delete.

Nice Work!

Nice Work!

You just built a complete data migration decision framework backed by a real, working DataSync implementation. You can now confidently evaluate any SAA-C03 exam question about moving data to AWS and recommend the right service based on volume, bandwidth, timeline, and compute needs.

You've learned how to:

  • Transfer data between S3 buckets across AWS regions using AWS DataSync, configuring locations, creating tasks, and observing the full execution lifecycle from Launching through Verifying to Success.
  • Calculate transfer times to identify the breakpoint where online migration becomes impractical, and apply the AWS Snow Family decision framework to choose between Snowcone, Snowball Edge Storage Optimized, Snowball Edge Compute Optimized, and Snowmobile based on capacity, compute, and portability requirements.
  • Navigate the Snow Family Console to understand job types (Import into Amazon S3, Export from Amazon S3, Local compute and storage only) and device selection options for real-world ordering workflows.
  • Design a hybrid edge-to-cloud architecture combining Snowball Edge Compute Optimized with DataSync for a manufacturing IoT scenario with 12 factory sites (Secret Mission).

Ready to quiz yourself?