Fastest Method to Transfer 10 TB Database Export to Cloud Storage

PCA By islever Editorial Last updated

Configuring individual storage systems. Creating a migration plan (i.e., documents and architectural diagrams).

Short answer

D Compress the database export and upload it to Cloud Storage using gsutil -m, completing the 10-TB transfer in roughly 30 hours at 1 Gbps, far faster and cheaper than Transfer Appliance.

Question

Your company wants to migrate their 10-TB on-premises database export into Cloud Storage. You want to minimize the time it takes to complete this activity, the
overall cost, and database load. The bandwidth between the on-premises environment and Google Cloud is 1 Gbps. You want to follow Google-recommended
practices. What should you do?
  1. Develop a Dataflow job to read data directly from the database and write it into Cloud Storage.
  2. Use the Data Transfer appliance to perform an offline migration.
  3. Use a commercial partner ETL solution to extract the data from the on-premises database and upload it into Cloud Storage.
  4. Compress the data and upload it with gsutil -m to enable multi-threaded copy. Correct Answer

Key Takeaways

The question tests Google's recommended data transfer practices for large datasets; the common trap is choosing Transfer Appliance for data over 1 TB, but at 1 Gbps a 10-TB transfer completes in roughly 30 hours—well under the ~20-day appliance turnaround—so gsutil -m is faster, cheaper, and still Google-recommended for this scale.

This PCA exam question asks the optimal method to migrate a 10-TB on-premises database export into Google Cloud Storage over a 1 Gbps link while minimizing time, cost, and database load. The correct approach is to compress the data and upload it using gsutil -m for multi-threaded parallel transfer, since the ~30-hour transfer time far beats the 1–3 week turnaround of Transfer Appliance.

Common mistake: Most learners choose B (Transfer Appliance) because Google associates it with large transfers over 1 TB, but at 1 Gbps the 10-TB upload finishes in about 30 hours versus 1–3 weeks for appliance shipping, loading, returning, and rehydration—making the appliance both slower and more expensive.

Expert Analysis

Why the Answer Is Correct

Option D—compress the data and upload it with gsutil -m—minimizes time, cost, and database load simultaneously. At 1 Gbps, 10 TB transfers in roughly 30 hours using multi-threaded parallel uploads, which is dramatically faster than the 1–3 week turnaround for a Transfer Appliance (ship, load, ship back, rehydrate). Google's architecture guidance on transferring large datasets explicitly states that Transfer Appliance should only be considered when the online transfer time is "substantially more than" the ~20-day appliance turnaround time—30 hours is nowhere close. Compression further reduces transfer time and the multi-threaded flag maximizes throughput without adding load to the database, since the export is a file-level upload.

Why the Other Options Are Wrong

Option A (Dataflow job reading directly from the database) would sustain heavy load on the on-premises database for the entire transfer duration, directly contradicting the requirement to minimize database load. Option B (Transfer Appliance) is designed for scenarios where the online transfer time substantially exceeds 20 days; here the online transfer completes in under two days, so the appliance adds weeks of latency and shipping costs with no benefit. Option C (commercial partner ETL solution) is not a Google-recommended practice and introduces unnecessary licensing costs and complexity for what is simply a bulk file transfer.

Community Comment Notes

The most upvoted comment by pr2web provides a concrete calculation: "Time to transfer using 1Gbps: 30 hours" versus "Time to transfer using Transfer Appliance: 1-3 weeks," concluding "Answer is D." Several learners like mikesp note that Google's guidance recommends Transfer Appliance for transfers exceeding 1 TB, but they overlook that the recommendation is conditional on online transfer time exceeding the appliance turnaround. Commenter moota correctly quotes Google's guidance that appliance consideration applies when online transfer is "substantially more than" 20 days, which does not apply here.

Official Reference

These links open the official vendor documentation in a new tab.

Exam Strategy

When comparing transfer methods, always calculate the actual online transfer time from the bandwidth given. If the online transfer completes far faster than appliance turnaround (~20 days), the online method (gsutil -m or Storage Transfer Service) is the correct choice.

Frequently Asked Questions

Why isn't Transfer Appliance (B) faster for a 10-TB database export at 1 Gbps?

At 1 Gbps, 10 TB transfers online in about 30 hours. Transfer Appliance requires shipping, loading, returning, and rehydrating data—totaling 1–3 weeks—so it is substantially slower for this volume and bandwidth.

Google recommends Transfer Appliance for transfers over 1 TB—why doesn't that apply here?

That recommendation is conditional: appliance is preferred only when the online transfer time substantially exceeds the ~20-day appliance turnaround. Here 30 hours is far less than 20 days, so gsutil -m is the faster, cheaper option.

More PCA FAQ →

Discussion 175 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 30 of 175

  1. pr2web 👍 108
    This is pretty simple. Time to transfer using Transfer Appliance: 1-3 weeks (I've used it twice and had a 2-3 week turnaround total) Time to transfer using 1Gbps : 30 hours (https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets) Answer is D, using gsutil
  2. PleeO 👍 23
    according to your real case, D definitely is a correct answer. However this is the GCP exam, I would go with B - Use data transfer appliance -> the precise answer
  3. gingerbeer 👍 21
    No perfect answer as B and D both have flaws. B is time latency as transfer appliance usually takes weeks; D gsutil applies for less than 1TB. The answer should be storage transfer service for on-premises data, which is not available here. If have to choose one I go for B
  4. mikesp 👍 20
    I agree. "Google-recommended practices" means than if more than 1TB, then Transfer Appliance. The main drawback of option D is that gsutil does not enable to manage bandwidth and corporate connection could be congested by transfer and affect other business activities.
  5. cloudguy2 👍 15
    B) Google recommends gsutil for <1TB; for >1TB, Storage Transfer Service is recommended. Since STS is not in the answers, the next best large transfer option would be offline Storage Appliance.
Show 25 more comments
  1. DeepDiveDevops 👍 8 Answered B
    D is not possible. max upload limit for an object using gsutil is 5Tb. https://cloud.google.com/storage/quotas
  2. joe2211 👍 8
    Not about time but "Google-recommended practices"
  3. moota 👍 7
    According to https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets > The expected turnaround time for a network appliance to be shipped, loaded with your data, shipped back, and rehydrated on Google Cloud is 20 days. If your online transfer timeframe is calculated to be substantially more than this timeframe, consider Transfer Appliance.
  4. Deb2293 👍 6
    Go home you are drunk
  5. zz05wh 👍 5
    The best option is B - Use the Data Transfer appliance to perform an offline migration. The key considerations here are: - Minimize time: Offline migration with the Data Transfer appliance will be faster than online extraction for a 10 TB database due to reduced load on the database. - Minimize cost: The Data Transfer appliance uses discounted network pricing, reducing overall cost. - Minimize database load: Offline migration eliminates load on the production database. - Recommended practices: Google recommends the Data Transfer appliance for migrations over 10 TB. The other options have downsides: A) Online Dataflow extraction adds database load. C) Commercial ETL adds cost, may be slower. D) Multithreaded gsutil upload will be bottlenecked by the 1 Gbps bandwidth. So for a 10 TB migration, the Data Transfer appliance provides the fastest, most cost-effective migration option following Google recommendations.
  6. Jay_Krish 👍 5
    The ask in the question is "minimize the time, cost and load on the DB" Option A : Not sure why no one talks about it? Is it not possible using Data flow? And that there's no downtime as the data is being read directly from the DB Option B : It says transfer offline.. one of the criteria of DB Load is not fulfilled as this means the DB has to be shutdown for whatever time (30hrs or 3 weeks as the discussion goes) Option D : Same as above I feel Option C: Ruled out as it is going for commercial tools. Appreciate if someone can throw some comments/discussion on this
  7. telp 👍 5 Answered B
    Answer B because it's a google exam. The rule for best practices is for more than 1 Tb to transfer data => use the google transfer appliance. But we all know in reality with 1gbs, we will use classic upload with gsutils. Ref. https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets#transfer-options
  8. sbgcp 👍 5 Answered A
    A is the correct ans. B) too expensive and maybe it will take more than 30h, Google suggest to use it for more than 100 TB. C) doesn't make sense. D) gsutil is recommended for less than 1TB of data https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets
  9. jay9114 👍 4 Answered B
    I agree with @mikesp. Based on google recommended practices the answer is B. In section, "options available from google", it states that you should use the storage appliance IF you (1) 1. Have enough bandwidth to meet your project deadline and need to migrate greater than 1 TB of data 2. Not enough bandwidth to meet your project deadline 3. Migrate data from another cloud provider to google cloud 4. migrate data from cloud storage to cloud storage Reference: 1. https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets
  10. Besss 👍 4
    Transfer Appliance is not correct. According to GCP documentation: The two main criteria to consider with Transfer Appliance are cost and speed. With reasonable network connectivity (for example, 1 Gbps), transferring 100 TB of data online takes over 10 days to complete. If this rate is acceptable, an online transfer is likely a good solution for your needs.
  11. kripya 👍 4
    Took the exam yesterday and gsutil option was not even there. Only 2 viable options were dataflow or transfer appliance. Question was to select most cost effective option, so went with dataflow. I did pass the test in first attempt.
  12. Sandipcst 👍 4
    -m means multi thread. It is useful for parallel uploads with large number of files, NOT large files like 10 TB. So, Option D will not work. Rather transfer appliance would be better choice
  13. VishalB 👍 4
    Answer D -- Question talks about minimize time, cost & database load. So Option D is more suitable for above scenario as it is using compressed files and parallel upload option
  14. joe2211 👍 4 Answered B
    vote B
  15. cloudmon 👍 3
    Unnecessarily difficult question that relies on a conflict between two Google "best practices" 1. gsutil is recommended up to 1TB of data 2. Appliance is recommended if your upload would take "substantially longer" than 3 weeks. In this case, the correct answer would be to use Transfer service for on-premises data, but they don't provide that option, thus forcing us to pick the best from a bad bunch of options, and hope that we luckily avoid whichever pitfall they are focusing on here (e.g., which of the above "best practices" will we not pick). They probably want us to choose the Appliance option (i.e., trapping us with the "up to 1TB" recommendation for gsutil), but I would never do that in real life. I'm voting for gsutil out of a matter of principal.
  16. ryzior 👍 3 Answered D
    I vote D: https://cloud.google.com/transfer-appliance/docs/4.0/overview Is Transfer Appliance suitable for me? Transfer Appliance is a good fit for your data transfer needs if: You are an existing Google Cloud Platform (GCP) customer. Your data size is greater than or equal to 10TB. Your data resides in locations that Transfer Appliance is available. >>> It would take more than one week to upload your data over the network. 30hrs is the point - you have to minimize the time, with transfer appliance it takes a week minimum.
  17. RitwickKumar 👍 3 Answered D
    Please note the keyword "Compress" in the option D. Due to lower entropy or randomness of data (which means similarity in values with in a column) would greatly reduce the size of the file from 1TB which means that despite the fact that the actual uncompressed data size is 1TB the compressed size should be much lower reducing the overall time to upload. gsutil with multi threaded upload of the file chunk fits the bill here and should not take considerable time.
  18. sank8 👍 3
    Answer is B Google-recommended practices" means than if more than 1TB, then Transfer Appliance. The main drawback of gsutil option is that it does not enable to manage bandwidth and corporate connection could be congested by transfer and affect other business activities. Reference: https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets
  19. MikeB19 👍 3
    This is the correct article to support this question but the article proves the transfer appliance is the correct answer. Right below the transfer calc chart is recommended amount of data for gsutil. Gsutil should be used for data transfer under 1 tb “Your private data center to Google Cloud Enough bandwidth to meet your project deadline for less than 1 TB of data gsutil”
  20. cert2020 👍 3
    Answer is B. Although the time taken for transfer of 10TB with 1 GBPS is 30 hours, Google recommends to use GSUTIL for data transfers less than 1 TB. Perhaps Storage Transfer Service would have been better choice but not listed. Ref. https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets#transfer-options
  21. PeterAH 👍 3 Answered B
    According to https://cloud.google.com/storage-transfer/docs/overview , a proposed question in the document is "Should you use gsutil or Storage Transfer Service?" , the answer states that Transferring more than 1 TB from on-premises - Use Transfer service for on-premises data. so I vote for B , please confirm if I am thinking correctly.
  22. lokiinaction 👍 3 Answered A
    I am thinking of A: for D transferring 10 TB with 1Gbps takes 3 hours: 10000/60/60 , but the problem is with -m option, database files are normally quite huge, which are not really good option to use multi threading Transfer Appliance is not required. Only meaninful option for me is A
  23. vladik820 👍 3
    I go with C. Google recommends use 'Storage Transfer Service for on-premises data' for 10 TB database. https://cloud.google.com/architecture/migration-to-google-cloud-transferring-your-large-datasets And its a Docker - commercial partner ETL solution. commercial partner ETL solution
  24. Nik22 👍 3
    10 TB is not much - 2 would be an overkill, D is recommended if it is upto 1TB, I don't understand, why to use commercial partner to move data, A also does not make sense since there is no data manipulation. I would go with D
  25. Sandipcst 👍 3
    Well - Option A is wining than D - because, to upload 10 TB data with 1 GBPS connection will take only 1 day. Whereas transferring through transfer appliance will take more than a month.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Multi-Regional vs Regional Cloud Storage for Global Software Downloads

S-grade

Surviving a Zone Outage With SQL Server on GCP

S-grade

Fixing BigQuery connection errors from a Compute Engine VM

S-grade

Moving 75 TB into Cloud Storage with Google-recommended practices

S-grade

Practice All PCA Questions

Access 81 questions with complete answers and detailed explanations.

View Full PCA Practice Test →

← Back to PCA Study Guide