The Data Center Followup

So, yes, it’s been a moment. Burnout is like that, though. But it’s time to hop back on this horse.

Despite that, I recently had the chance to talk at JiraCon ’22. Let me take a moment to thank Patrick, Ashley, and everyone at Trundl for the opportunity. It absolutely made my day to see people excited to see me, calling me out in chat. And to the person who said they learned so much in the presentation – thank you. Dopamine straight to the system, right there.

So, what did I talk about? Well, that’s what I was hoping to cover today. It’s been a few months since I first asked, “What exactly is Atlassian’s plan for Data Center?” Of course, we got an initial answer, to which I already responded, but so much has happened since then. And I promised you a followup once all was said and done, so…here we are.

We’ve had an AMA – which being completely honest, felt more like an advert for Atlassian Cloud Products. Atlassian, when people are there specifically for Data Center, constantly saying, “But that’s available in Cloud,” doesn’t do much to ease our concerns about the platform we’ve already chosen despite Cloud.

My team also got the chance to speak with Atlassian’s Head of Data Center Products, which was informative in and of itself. It was a few months after the AMA, and I could already see the needle start to move. That meeting didn’t completely aswage my concerns, but it was nice to see where things are going.

So, what’s the future of Data Center? Let’s take a look at that question right now.

Predicting the future is hard.

First, let me say that – in spite of what some would believe – I don’t work for Atlassian in any respect. Therefore, I’m not privy to any secrets that aren’t publically available, and I don’t have insight into Atlassian’s internal discussions. In fact, given my pension to write what’s going on in the Atlassian community, I get the feeling they are cautious with what info they give me – even in my capacity as a practicing Jira Admin. 

Given this blind spot, predicting the future is hard. History is littered with people who tried to predict the future and got it wrong. Occasionally, this fact has led to the downfall of otherwise profitable businesses. My favorite example is this from the movie rental business Blockbuster, whose CEO stated in 2008:

“Neither RedBox nor Netflix are even on the radar screen in terms of competition,” – Blockbuster CEO Jim Keyes, 2008

Yeah. Between RedBox and Netflix, Blockbuster filed for bankruptcy in 2010 – just two years later. 

Another of my favorite goofs is from IBM, who, in 1943, stated: 

 “I think there is a world market for maybe five computers, “– Thomas Watson, President of IBM, 1943

I think I have more than that in my house alone, especially if you consider things like smartphones, tablets, and IoT devices. But, I think we made our point. These quotes were made with the best info these people had at the time, and they still got it wildly wrong.  

So now I’ll try to predict the future.

Atlassian has been clear on one stance: Data Center isn’t going anywhere. To a point, I agree with them – but I’ll get to that later. According to Atlassian, they intend to carry on Data Center into the foreseeable future. Now, it won’t get as many resources as Atlassian Cloud products, which we’ll see as fewer features that inhabit a smaller scope. Atlassian has also been clear on this.  

More specifically, Atlassian intends to focus on three key areas regarding Data Center. 

  • Security Updates
  • Performance Improvements
  • Quality-of-Life improvements for Admins

I haven’t been the only one to mention that two of these three cover the bare minimum you would expect of a “supported product” – not something you should be holding up as this “great thing we are doing.” 

This leads me to my opinion: I also think that Data Center will be here to stay for the better part of a decade, but don’t count on it being around forever. I’ve listened to their points, and read everything they’ve put out and I don’t see anything to change my mind.

Let me be clear; I don’t think Atlassian is lying either. “How can that be?” I bet you are asking. Well, the Devil is in the details – or in this case, the words “Foreseeable future.” So that does put a bit of a cap on things. If they suddenly change plans and announce that Data Center will be sunset, that was due to unforeseen circumstances.

HOWEVER, I don’t think we all need to panic yet. There are two reasons I don’t think any announcement regarding Data Center is imminent. First, way too many organizations depend on it. Not just “Cloud isn’t certified in this specific thing, so we can’t use it yet.” No, people who have a literal legal requirement never to have their data in the Cloud. Again, I’m reminded of the air-gapped U.S. Department of Defense Jira systems I constantly hear rumors about. 

But Server was also in the same boat, wasn’t it? Well, not really. People still had a locally hosted option if they had to move from Server. It wasn’t cheap – by any means – but it was available. But even with this, there is another reason I don’t think Data Center is going anywhere.

In a word: Server. Until it was announced, Atlassian claimed they had no plans to Sunset Server. The way they went about it seriously hurt their credibility in many people’s eyes, myself included. They know this, which is why they responded so fast when I started asking questions. Atlassian also knows that if they suddenly announce another reverse course with Data Center, that might worsen the breach of confidence. Given that, I don’t think they are in a hurry.  

That said, I don’t think Data Center will be around forever. First off, everything has an end. It’s just the way of things. But secondly, this is the way the industry is moving. Atlassian’s newest competitors are cloud-only. And being Cloud only lets them move fast. So if Atlassian is to compete long term, they know their future is also in Cloud. I can’t say that makes me too happy (I still read “Cloud” as “Someone else’s Server”), but it is what it is.

Now, if I had to put a date to it, I foresee it being 5-7 years before Atlassian announces anything on Data Center. This timeframe gives them time to complete the sunset of Server (Feb. 2024 is coming fast) and some time to recover their credibility.  

That is still plenty of time, though. Even if I’m correct, that’s a long time before you worry about bulk Data Center migrations. So don’t stop taking those Data Center Certifications, taking jobs at companies that use Data Center, and using the platform. But maybe… keep an eye on the Cloud side as well. As they say, “Luck favors the prepared.”

So, what preparations should we make?

I’m glad you asked. Thankfully, the early prep you would make for a Cloud Migration are things that can help your instance run better.

So the first thing I’d do is take stock of what you have. Ask the following questions about your config:

  • Who is using this?
  • How important is each item?
  • Can I get rid of any?  

Typically, I go through these questions for each object in these configuration categories:

  • Apps
  • Projects & Spaces
  • Workflows
  • Custom Fields
  • Permission Schemes
  • Issue Types

Now, this will take a fair bit of effort and detective work on your part to answer these questions. But it will help guide you as you start a cleanup.

There are tips and tricks to discovering ownership in cases where it’s not clear for an App or Project, but I intend to gather these and do a separate article on them.

So once you’ve taken stock, you need to plan a cleanup. Of course, a cleanup will help in a migration, as everything you remove is something you don’t have to worry about moving. But even if you never have to migrate, it will still help your instance perform better, making this a worthwhile activity. I’ve done some articles in the past on how to clean up various aspects of your Jira instance, but I want to cover a newly discovered resource.

While doing some work for well, work, I found a repo of Groovy scripts that will help you clean out unused objects like workflows, issuetypes, schemes, etc. If you have Scriptrunner, I’d seriously suggest checking these out.

However, there is also newly published guides from Atlassian on this topic. This is a topic I know I’ve asked them about before, so it’s nice to finally have something from Atlassian in hand to help with this task. It’s definitely an interesting read, so I suggest you check it out.

So what do you think?

Are you convinced Data Center is staying? Are you preparing to move to Atlassian Cloud or some other platform? Let me hear you!

You can always find all my social media links on my Linktree. Be sure to like and comment on the social posts. You can also subscribe to the blog directly to get new posts in your inbox. Just put your email in the box below!

But until next time, my name is Rodney, asking, “Have you updated your Jira issues today?”

4 Comments

  1. Thanks for your blog, Interesting read always!
    Now on Jira , epic battle for Atlassian customers – cloud Vs DC.
    Point is Jira DC customers are usually large corporations , min < 5000 users and can go upto 20,000 users. So we are talking about scale and investment of many years by that corporation and also THIRD party apps atleast for last decade. I will not be lying if many organization have their first Agile and then SAFe transformation through JIRA. Now do we have all this in Jira cloud, before that, is their migration tool ready (it fails in exporting even workflows) , many features are not working. Ok Leave Jira , what about Apps , most of the apps are not ready for Atlassian adventure, they are struggling to offer what they offer in in-premises. So in the end at this scale of 10,000 users, 10 years of investment on in-premises JIra- we are NOT talking of migration of same product like A(DC) to A (CLoud), infact its A to B (only name is same! ). And If its A to B and requires lot of re-engineering then why stick to Jira, why not go shopping, why not something else at half pricing !!!

    Like

    1. You already answered the question yourself on going shopping : “So we are talking about scale and investment of many years by that corporation and also THIRD party apps at least for last decade.”. Remember the time it took to become an Expert on a subject? Not many companies are willing to go through that curve again. And while you might complain about migration assistant it does a fairly good job (specially compared to other suppliers).

      Liked by 1 person

      1. You like it or not, we are on that curve again as Jira on cloud is different from Jira in premises (from integrations and custom work done, point of view!) . On migration assistant, it is improving but its not seamless . List of what is not migrated is getting shorter but will not be zero -https://support.atlassian.com/migration/docs/what-gets-migrated-with-the-jira-cloud-migration-assistant/

        And third party apps maturity and migration is altogether a different mountain to climb. Let me take ScripRunner example – Their official page says
        “ScriptRunner for JIRA Cloud is a product that is under development and investment is being made to add features to the platform. It doesn’t currently have the same feature set as the server version and some features are likely never to come to Cloud due to restrictions in the Cloud platform.”

        Like

  2. Thanks for linking that Cleanup Guide! Found things in there to add to my own list.

    Hmm, that reminds me, I should make a cleanup guide for my fellow admins where I’m at… regardless of whether Cloud is in certain folks’s future, a clean Jira is a happy Jira.

    Like

Leave a reply to Anurag Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.