does this mean that return of the IT generalist?
What? sorry I must have nodded off. I have always been an IT generalist. What the future brings we shall see.
Hard is the future to see. An IT generalist must have the deepest commitment.
Mid-sized companies can reduce technical debt and free up IT budgets by switching to hyperconverged infrastructure systems (HCIS) in their next data centre refresh. So says Gartner, which defines 'mid-market' as companies with fewer than 200 virtual machines. The IT analyst firm's recommendation appears to be getting through …
Also if you have the less than 200 VMs then you are probably are not large enough to have specialised teams, if you do then there is one place to cut costs. Larger companies, with thousands of servers, 10s-100s of storage systems, networks crossing the globe, would have specialised teams, the type of companies that they don't say should use these.
I had a manager who regarded Gartner report in much the same way St Joan did the voices in her head.
Our actions were run to the dictates of these divine utterances and even if it lead to us being burned at the stake we could meet our doom face-on, consciences clear in the knowledge that we had done the Right Thing.
Yep. The components sprawl out as we pick-a-part to accelerate each bit according to our particular loads. And then we stuff it all in a chassis (spread across racks side by side if necessary) when latencies becomes an issue. "As the IT World Churns...."
Everything I've ever seen to date has looked like something found in the IBM 360's and 370's I learned on as a kid, from IBM engineers as a matter of fact (with a heavy helping of manuals). There have been oddball things done but still.... So, looking around, rack to datacenter, all I see is mainframe.
Personally, I believe that if your in this biz for the long haul, you have to be able to operate as one or more kinds of specialist and a decent generalist. So Gartner is simply describing a simple truth. And making a ton of money doing it.
Don't get me started on the infrastructure's team's predilection for putting app servers, database servers and storage as far apart as possible. Seems their ease of maintenance trumps the application actually working.
The worst I've seen is a data warehouse in one country, and the ETL servers in another country - and not even an adjacent country at that. I guess they could have been in different continents ...