There was a well thought-out post on SystemCenterCentral.com that I wanted to respond to. I’ll quote out the post, and respond inline. I’m doing this here because it’s a topic near and dear to my heart, and one I want to be able to easily refer people to in the future.
As someone who has invested a significant amount of time learning the System Center product suite, I am beginning to wonder if focusing on System Center is a good move career-wise.
It’s pretty clear that the importance of DataCenter management tools is lessening. The trend is for companies to move to the Cloud and leverage SAAS and IAAS solutions.
Well, yes, but also no. Private data centers will remain hugely important for a long time, and you’ll need tools to manage them – as well as tools to manage your public-cloud assets. It’s going to be a hybrid world for the foreseeable future. Many organizations simply won’t go heavily public; there are a lot of political and legal concerns around it right now that aren’t going to be resolved real soon.
The prevalence of On-Prem tools is diminishing and everything is moving towards a service-based model.
Well, no. Your own datacenter should have been moving toward a service-based model – e.g., a service you provide to the organization – a long time ago. You need on-prem tools to do that. One reason the public cloud has been successful is because so many internal IT teams utterly failed to provide their organizations with what the organization really needed, often due to a lack of executive leadership.
How many Sharepoint and Exchange admins do you need when you have Office365?
Fewer, for sure, but O365 isn’t admin-free. Frankly, most organizations I see are heavier on communications admins than they need to be. It’s O365’s automation that lets you get away with fewer admins; it’s automation you could have implemented internally some time ago.
Why stick with System Center when you have free Open-Source solutions such as OpenStack available?
The same reason you use Windows instead of Linux. Open-source isn’t the perfect solution for every organization. Many organizations will find benefit in having multiple stacks serving different business purposes.
Why invest in SCOM monitoring when you can just use AWS and leverage AWS Cloudwatch which is free?
Because you’ll still monitor internal assets, and because Cloudwatch doesn’t monitor to nearly the level that SCOM does. Further, you’ll need to monitor hybrid assets, and something like SCOM is well positioned to do that if your public cloud assets happen to be in Microsoft’s public cloud. Yes, System Center is a proprietary stack – it’s pretty much always been that way. Everyone else’s stack is proprietary, too; “Open” Stack might be open-source, but it still wants you to implement its components rather than someone else’s. I don’t mean that as a bad thing, just pointing it out.
Why use SCCM to patch servers when you can slipstream patches into your server build scripts with CloudFormation scripts?
Personally, I question whether you’ll use SCCM for anything for much longer, as I think better solutions are coming. But, this isn’t a “does System Center offer any value” question, it’s a “whose stack do you want to play in” question. If you’re implementing a Microsoft stack, and that’s inclusive of Azure-based assets, you’ll use Microsoft tools. This question is like saying “why would you use SCVMM when vCenter does the same thing?” The obvious answer is, “because they’re different stacks.”
Why use Orchestrator when you have SMA?
You probably wouldn’t. Orchestrator isn’t Microsoft’s go-forward solution; SMA is. But SMA is part of System Center; this doesn’t in any way point to System Center being less relevant.
Why bother deploying SCSM when you can leverage a SAAS solution like ServiceNow and take the complexity of owning and operating a helpdesk application out of the equation?
Again, this is a stack question in part. SCSM is the hub that, along with SMA and Orchestrator, tie most of the Microsoft stack together. And SCSM isn’t terribly complex to own or operate. I also expect Microsoft will someday rent us a SaaS version of SCSM (and likely SCOM, along with other bits). You have to be a bit careful about making this an “on-prem vs. cloud” argument, because System Center is edging toward being “cloud” itself.
Look at what Microsoft has been doing: They ended Technet subscriptions. They ended MMS and have combined all of their conferences into one huge uber-conference. They ended the Master MCSE certifications.
This doesn’t really have anything to do with the System Center question; this feels more like a “is Microsoft abandoning IT pros?” question. And yes, in part, they are. The days of clicking next-next-finish to manage your infrastructure are ending; DevOps is what Microsoft wants running the datacenter now, and that means admins who think a bit more like developers… and that means ending the separation between “dev” and “IT pro.” You don’t have to like it; you can certainly point your career elsewhere if you think Microsoft is off it’s rocker. But this is what they’re doing.
The Master MCSE certifications were almost always doomed – few people took them, they were expensive, and they just weren’t following the business trends. They were, in some ways, a day late and a dollar short to get really entrenched. Do you have one? I never did.
TechEd, MMS, and the other conferences were, in my opinion, just a stupid marketing move. I don’t think they’re part of any bigger pattern, apart from little fiefdoms inside Microsoft who were getting agitated that other people were running conferences.
Everything they are pushing now is Azure Azure Azure. Microsoft is transitioning into becoming a service-based company.
True, and they’ve been very upfront about that with phrasing like “cloud-first engineering.” Microsoft’s customers drove them to it, in part, by sitting on a single server operating system for over a decade. If Microsoft can’t sell you software in a steady stream, they’re going to rent it to you instead.
Look at their latest System Center technical preview: SCVMM now no longer supports Citrix XenServer and VMWare:
Which is hugely disappointing, but not unexpected. You’re right in that the company is pushing for what they need in Azure, and the SCVMM team doubtless felt pressured to drop “outlier” support and focus on core stuff.
The bottom line? The Microsoft IT admin career path is changing. That doesn’t mean System Center is going away, but it’s going to morph. Being a guru is probably still a safe career path, provided you keep up with the changes. Don’t tell me you’re a guru in SC2007; nobody cares. But if you’re solid on all the latest – and the change is going to be fast and drastic from here on out – then you’ll be of great value to many organizations. But, if you’re just the guy who runs the Deployment Wizard in SCCM… yeah, I’d start looking at other futures.