Reduce the cost using AWS S3

[Kei]

Hi

Now I use VBR(ver 12) installed on AWS EC2 and EBS as backup repository.
But EBS has the storage capacity limitation, so I would like to migrate to AWS S3.

I calculated the S3 cost roughly and I noticed S3 cost will be quite high rather than EBS because of the API call to S3.
(Regarding to past answer on this R&D Forums, 1 million API call will generate per 1 TB backup.)

Also I considered to use S3 Glacier as Archive tier using SOBR, but the cost is also quite high rather than EBS because of the API call to S3.

Is there any way to use S3 cheaper?

Best regards.

[Mildur]

Hi Kei

If you want to save costs on API calls, you can change the block size of a backup. Default is 1Mb, you can go up to 8MB.
Bigger block size, fewer API call, but incremental backups will be bigger (1.5-2.5 times bigger with 4MB block size).
Changing block size on existing jobs require an active full backup.

https://helpcenter.veeam.com/docs/backu ... timization

Best,
Fabian

[Kei]

Hi Fabian

Thank you for your reply.
In the past post, I found the default block size provides the best balance between API and storage costs.

object-storage-f52/aws-s3-how-to-reduce ... 68858.html
+++++++++++++++
For S3 Standard, the default block size usually provides the best balance between API and storage costs. Using large blocks may both increase and reduce your overall bill, this depends solely on the type of protected data. Because some workload types may results in very large incremental backups when using a large block size (if they create large number of small changes across the entire disk).
+++++++++++++++

I would like not to reduce the cost of API calls, I would like to reduce the cost of S3 as a whole.
Is there any way to reduce the cost of S3 as a whole ?

Best regards.

[Mildur]

Hoi Keita

Thank you for the link.
I assumed something in this direction, that's why I noted the higher storage usage for incremental backups.

There is also S3 Standard-IA, but I don't think it will be cheaper. S3 Standard-IA is for infrequently accessed data. If you move your backups to Archive Tier, you will have frequent access to backups on your AWS S3 buckets. You can save money on storage, but API Calls are much higher.

Cloud is not cheap. I don't see options to save a lot of money on AWS. If you want to save money, have you checked out other providers without API calls like Wasabi? If I remember correctly, your backup server in AWS manages on-premises agents from your company. Direct Backup to object storage is possible with such scenario.

Best,
Fabian

[Kei]

Hi

Thank you for your reply.
I understood. I would like to compare other cloud object storages.

Best regards.

[Kei]

Hi

I am sorry for sometimes asking. One additional question about using S3.
If I use SOBR as bellow tier, Performance and Capacity tier needs additional API call to move or copy data to next tier?

Performance tier: Amazon S3 standard
Capacity tier: Amazon S3 standard
Archive tier: Amazon S3 Glacier

Performance tier: 1 million Put request per 1 TB and 1 million Copy request per 1 TB
Capacity tier: 1 million Put request per 1 TB and 1 million Copy request per 1 TB
Archive tier: 1 million Put request per 1 TB

Best regards.