Container Selection Plan: azure-blob-poc-selection-us-east¶
Container Analysis Summary¶
| Container Name | Region | Total Size | Blob Sizes | Blob Count | File Types | Permissions Complexity | Notes |
|---|---|---|---|---|---|---|---|
| finance | eastus2 |
~41.2 GB | Mix (KB to 100MB) | 4,285 | Backups (.bak), SQL, XML, Docs | High (Net Rules) | Strongest candidate. Good size mix and region alignment. |
| engineering | eastus |
~1.02 GB | Mostly Small (KB/MB) | 4,525 | Dev artifacts (.whl, .parquet, .js) | High (Net Rules) | Good secondary. Excellent for small-file performance testing. |
| human-resources | centralus |
N/A | Mix | 4,150 | Docs | High | Wrong Region. |
| legal | centralus |
N/A | Mix | 3,980 | Legal Docs | High | Wrong Region. |
| marketing | canadaeast |
N/A | Mix | 3,860 | Marketing Assets | High | Wrong Region. |
| product | canadacentral |
N/A | Mix | 3,894 | Product Specs | High | Wrong Region. |
| sales | canadacentral |
N/A | Mix | 4,247 | Sales Data | High | Wrong Region. |
Recommendations for proof of concept migration¶
For the Proof of Concept (POC) migration to the US EAST region, the finance container is the optimal selection.
Primary Selection: finance¶
Reasoning:
- Region Alignment: Located in
eastus2, which is a primary US East region, minimizing latency and egress costs during the POC. - Data Representative:
- Size & Scale: With a total size of ~41.2 GB, it provides a substantial dataset to validate throughput.
- Object Mix: It contains a realistic mix of "larger" files (Database backups ~100MB) and smaller documents (PDFs, XMLs). This variation is critical for testing multi-part upload settings and throughput scaling.
- Complexity: The container enforces strict Network Rules
(
defaultAction: Deny), providing a necessary test case for mapping Azure VNet/IP restrictions to Google Cloud VPC Service Controls or IAM conditions.
Secondary Selection: engineering¶
Reasoning: The engineering container (eastus) is a strong alternative
or add-on. While smaller (~1GB), it contains a high density of small files
(code, logs), which is excellent for stress-testing "metadata-heavy" operations
and evaluating object creation latency (Requests Per Second) rather than just
bandwidth.
Conclusion¶
Start the POC with finance to validate throughput and large-file handling, optionally adding engineering to test small-file performance.