Just to add nothing like cloud watch is setup :) 

I never used such a thing

Why do you want to do this? Wouldn’t it be easier to use cloud object storage?

Not me a customer. They want a method to long term Quarterly archive from a Tenant repository on Cloud Connect. For tape jobs there still exists Quarterly jobs unlike with Backup copy jobs. 

I’m interested in this as well. Cloud pricing is such a mystery to me. 

Looking into archiving some data, but no one can tell me what kindof retention policies or how much would be retrieved so it’s impossible to quote.  At least physical tape is great for things like Quarterly jobs you can export.  One day I’ll convince them for immutable object for copy jobs. 

If it’s anything like Azure, I'd assume they charge you though. lol. 

The cloud tapes are nothing different ithan object storage n the background …

I would probably go with Cloud storage, object.
You can configure it as immutable for long time.

For the costs / maths, its true, you don't know exactly how much you will Pau with storage, egress fees, etc…

you know that if its cost vs amount of data, wasabi can help.

there is many workarounds nowadays, will depend on what you relay / stand for, and also your customer decision and likes.

1. Situation is a given, I have no choice in the matter.
2. There must be Quarterly Archives (these are no longer present in backup copy jobs)
3. This is Cloud Connect (it is a tenant repository)
4. AWS Tape Gateway is installed and connected to a tape server that is in turn connected to Veeam
5. Does anybody know how often Veeam’s tape setup will leverage AWS apis even when not performing a tape job? I only ask because when Idle I see it quite often doing a rescan (so not the daily run that you set). Thanks 

Will be interested in this topic as being an AWS Community Builder I have credits for AWS to test with so seeing this evolve will be good.

No speaking for tape or VTL specifically, but I’m also having a hard time estimating API’s calls and ingress/egress as well.  I’m not much of a cloud guy when it comes to the compute side of the house, but I’m trying to learn it.  Most of our clients, now all of our clients operate their VM’s on-premise, so learning about building new VM’s in Azure or lift-and-shifts is a huge learning curve for me.

Hi Geoff,

I haven’t got the exact breakdown off efficiency vs object storage, but allow me to help set some expectations that can be overlooked when reviewing this option:

• Object storage doesn’t have a concept of full backup vs incremental due to the individual objects, which will give you much better storage efficiency vs tape if you’re going to be uploading entire full backups every 3 months to individual virtual tapes.
• When restoring one of these points, with object storage if you’ve got say 75% of the data identical to a local backup, you’ll read the 75% locally and just incur egress bandwidth, and API calls for reading the 25% remaining data, this also makes the recovery faster. Compared to VTL where all the blocks must be read for a restore IIRC. This amplifies in cost dramatically when you work with colder storage too.
• Fully understand the frustration of no quarterly GFS anymore, but it might work out cheaper to use monthly’s to cover the gap rather than VTL, depending on data change rate.
• As far as pricing goes, AWS don’t charge the same for tape gateway vs S3, they charge purely via data read & written in GB per month

Check out the AWS calculator and you’ll see there aren’t any references to API billing for the storage consumed. But I’d definitely test in the lab if there are hidden costs not shown in the calculator…

This is the part that kills me….trying to find the hidden costs, and this is what makes it hard for me to try and tell a client how much things will cost.  Some of it is a shot in the dark.  And since I have no experience with such things, they get to be my Guinea pig, and I have to do my best to make an educated guess until I have a bit more experience to rely on.  I’m sure that’s the case for everyone, but it drives me crazy to try and figure this part out.

Thanks for the advice guys but I can’t change this setup it is not mine I can’t even go and influence and changes (long story, not a business organization and under a lot of compliance rules).

Again this is a Cloud Connect Tenant which means that apart from manually picking and choosing files and uploading them to S3 (maybe some kind of ugly script), I personally don’t know of any way to automatically grab files from the tenant repo other than Tenant to Tape Backup in Cloud Connect, which is officially supported.

Either way I was just wondering if anyone had done AWS gateway and the api calls. @MicoolPaul thanks for that calculator, that is a great start. I might ask Veeam as well. Dustin has a whitepaper so maybe he saw this.

Sorry I couldn’t have been of more help there! Have steered people away from that since at least v10 so I thought I’d just share the considerations for if you still have some sway in the design. Be sure to report back please

No probs what you gave me was perfect as I can say “use the calculator” and not have to commit to an answer :)