Cloud or DC?

[Dominika Kuraś-Moskwa [Deviniti]]

Does anybody else hear the clock ticking? The Server end date is closer and closer, and I am wondering... team Cloud or team Data Center? And Why?

[CDS Administrator]

Dominika -

We were running Server (on prem) and went with Data Center.
Pros: no conversion or migration. It was a simple matter of applying a license key.
Cons: Minimum license is 500 users. We have less than half of that. So it cost us a bit more.

[Dominika Kuraś-Moskwa [Deviniti]]

Thanks for your input!

[ITMAGE]

We are using the on-premise SERVER product and would certainly entertain moving to Datacenter, but the costs (at least for us) are just too phenomenal. Moving to the cloud is not an option for various reasons including one of principle, and so after much consideration and the fact that our instances are not Internet facing, we have decided to remain on the server product indefinitely. The server product has been reliable, stable, immutable and is well understood in it's capabilities and operation. The cloud on the other hand is another matter, a black box, and like pretty much all cloud providers, would put us directly at the mercy of the Atlassian and the app vendors. This means there could be forced changes to the UI or a feature/operation, potentially forced updates to the various apps we use changing their operation (and what if one of those app providers ceases business, will the app be pulled due to lack of security updates for example?), questionable reliability, potential for some unrecoverable issue resulting in data loss and a myriad of other undesirable scenarios. In contrast with the local instance, it's under our direct control and we have clear policies to deal with most all scenarios. I am sure there are those who would question the wisdom of running with unsupported software with no future patches, but that simply is not an issue or major concern in our specific instance. I am not recommending others do the same, but for us this is the best option going forward. You know,,, the developers at Atlassian did a superb job with building Jira and Confluence on-premise server (and I very rarely give high praise) - you should all be very proud of your achievements. And I am sure there are those at Atlassian who wish to continue the on-premise products, but I guess those in upper management are trying vehemently to move everyone to the cloud, and the pricing of DC really conveys that message in my opinion.

[Dominika Kuraś-Moskwa [Deviniti]]

Thank you for your answer! 

[John Dunkelberg]

This will vary a lot based on your company's functional needs, size, and budget.  I'm guessing you're a smaller company since you're still on Server, so some of my experience/opinions may not apply.  Bottom line is that I think you largely have a short-term vs long-term decision ahead of you.

At my company, we're on DC and checking in on Cloud on an annual basis every Q1.  The functional gaps are closing, and depending on what your company has done in the realm of customization there may be no functional gap at all to Cloud.

The cost of Cloud is higher, though you certainly should factor in your in-house costs to maintain your Server (and potential DC).  And yes, for many companies this is not a good time to go to the budgeting process and say you need a substantial increase in costs. :)

Atlassian has been clear about being cloud-first, and they've demonstrated that in terms of the roadmap for the products.  If you go to DC now, are you going to be revisiting a migration to cloud in 2-3 years?  Long term, are you going to cost your company more in terms of disruption of service and in-house effort to move to DC now (fairly painless) and then move to Cloud later (disruptive, expensive) or to pay the higher cost starting now.

But that's going to be driven by your local requirements and constraints.

[Dominika Kuraś-Moskwa [Deviniti]]

Insightful! Thank you!

[David Cowley]

Both. We're doing the hop to Data Centre first so that we're never on unsupported software. We've got about 10,000 users on Jira and Confluence. 100+ projects, 250+ spaces. We can't complete the migration to Cloud Premium before February, so we're jumping first to DC, then taking the year to do a phased migration to Cloud using dual licensing.

Cloud aligns with our company direction overall. Personally, I'd be okay with stopping at Data Centre and there's some privacy and security assessments that need to be complete to finalize the move to Cloud.

There's a separate initiative within the organization to go to Cloud Enterprise instead of Cloud Premium but that's beyond the scope of our migration project. Our main concern is being on supported instead of legacy software, and ensuring the continued ability to use all existing projects indefinitely.

We have other instances of Jira and Confluence within the organization which are not externally accessible that are intending to upgrade to the last LTS and stay there indefinitely.

Our timing and decision making are largely related to the funding process. If funding wasn't a consideration, I'm not sure where we'd end up. But it's easier for us to go all the way to Cloud Premium than it is to get approval for two migration projects in different years.

The move to Cloud Enterprise if it happens would include consolidation of other instances of Jira and Confluence which are beyond the scope of the current project.

[Dominika Kuraś-Moskwa [Deviniti]]

Thank you for your answer!

[David Cowley]

Actually, we just heard of a new program that we believe is available to us that would prevent us from needing to do the hop to Data Centre. We don't have confirmed details yet, but it seems as though if you have a 1,000+ user count license that you may be able to do dual licensing with Server and Cloud instead of Data Centre and Cloud. If this pans out, we would likely take that approach instead of moving to Data Centre at all. It extends Server support to Feb 2025.