Skip to main content
  • EN

Hello,

One of our customers uses VBRv12 on premise with Azure an Repository. They have about 2TB of data backed up locally and via backup copy jobs to an Azure Blob Repo.

We were very baffled to notice that at the end of the month, the cost with transactions was matching the cost of the storage.

I am looking for an advice on how to optimize the number of transactions other then reducing backup frequency.

For a fact we checked his environment and he has 1server per job and 1 backup copy job per production job. I wonder if aggregating all in a single backup copy job towards Azure would address maybe this issue?

We are also testing this scenario but to be relevant we need to run it for at least on week so if anyone ran into this please let me know.

If you have many copy jobs that would be the transaction part of your query.  I would suggest trying to combine things in to one copy job (depending on the number of VMs) and see how these compares.  If need be, you can have two jobs, but many could cause the transactions you are seeing.

I have a little update here: we tested internally but it seems unfortunately that the number of backup copy jobs does not affect the number of transactions, but rather the number of the original jobs from which the VMS/Physical server are selected for the backup copy job. We will continue testing and let you know.

Hi, two questions here:

1: What is the block size configured on the job?

2: Hot or Cool tier Azure storage are you writing to?

Hi, two questions here:

1: What is the block size configured on the job?

2: Hot or Cool tier Azure storage are you writing to?

Hello, good questions:

  1. Default, I think it’s 1MB.
  2. Hot storage, we made some calculations and while cool tier storage is cheaper, the price per transaction is way higher. 

Hi @Catalin Adiaconitei double check this please as I wrote a blog post on API costs vs backup storage a while ago: https://www.veeam.com/blog/cloud-object-storage-benchmarks.html and what you’ve described would make sense with either a smaller block size, or utilising less expensive storage due to the higher API costs. Please confirm and let me know!

Hi @Catalin Adiaconitei double check this please as I wrote a blog post on API costs vs backup storage a while ago: https://www.veeam.com/blog/cloud-object-storage-benchmarks.html and what you’ve described would make sense with either a smaller block size, or utilising less expensive storage due to the higher API costs. Please confirm and let me know!

Thanks for the heads-up, I will look into it.

Comment


Terms & Conditions