You are absolutely spot-on—the egress fee is the definition of a silent budget killer! It's the point where a robust plan becomes completely unaffordable, which is why most of us have to accept the Active/Passive model over Active/Active.
To your excellent question about using object storage for DR backups instead of full database replication: Yes, that is exactly the strategic compromise most organisations should be making.
The Object Storage Trade-Off
Using object storage (like AWS S3, Azure Blob Storage, or Google Cloud Storage) for your cross-cloud DR backups offers massive advantages:
1. Cost Efficiency: Object storage is dirt cheap compared to highly available, instantly recoverable database services. This significantly cuts your overall standing DR budget.
3. Vendor Neutrality: Object storage APIs are very standardized, reducing the proprietary lock-in we discussed. It's often the easiest way to move large volumes of data across clouds.
However, we must be clear about the trade-off:
While object storage validates your RPO (you know the data is safe and recent), it significantly impacts your RTO (Recovery Time Objective). If you have to recover, your process now includes the time needed to:
Download the data from the object storage.
Provision the database service on the new cloud.
Ingest or restore the data into that newly provisioned database.
This process can take hours, meaning you accept a longer RTO for those critical systems in exchange for massive cost savings.
It’s all about balancing the risk (a longer RTO) against the reward (a manageable budget)! Thanks so much for highlighting this practical and important distinction!
You are absolutely spot-on—the egress fee is the definition of a silent budget killer! It's the point where a robust plan becomes completely unaffordable, which is why most of us have to accept the Active/Passive model over Active/Active.
To your excellent question about using object storage for DR backups instead of full database replication: Yes, that is exactly the strategic compromise most organisations should be making.
The Object Storage Trade-Off
Using object storage (like AWS S3, Azure Blob Storage, or Google Cloud Storage) for your cross-cloud DR backups offers massive advantages:
1. Cost Efficiency: Object storage is dirt cheap compared to highly available, instantly recoverable database services. This significantly cuts your overall standing DR budget.
3. Vendor Neutrality: Object storage APIs are very standardized, reducing the proprietary lock-in we discussed. It's often the easiest way to move large volumes of data across clouds.
However, we must be clear about the trade-off:
While object storage validates your RPO (you know the data is safe and recent), it significantly impacts your RTO (Recovery Time Objective). If you have to recover, your process now includes the time needed to:
Download the data from the object storage.
Provision the database service on the new cloud.
Ingest or restore the data into that newly provisioned database.
This process can take hours, meaning you accept a longer RTO for those critical systems in exchange for massive cost savings.
It’s all about balancing the risk (a longer RTO) against the reward (a manageable budget)! Thanks so much for highlighting this practical and important distinction!