
So much fluff...
In that fluff piece...
Extra points for using "square that circle".
Still not in enterprise infrastructure so its nice to read about how it can be done.
Thanks to Gartner, we have a new buzzword: bimodal IT. It’s nothing special actually, just a new way to describe common sense, and the fact that the world – the IT world in this case – is not black or white. In practice, in modern IT organisations it is better to find a way to integrate different environments instead of trying …
"it is possible to collapse different infrastructure stacks onto a single larger infrastructure capable of serving all physical, virtual and cloud concurrently"
Is it just me, or has anyone else's bullshit bingo starting ringing ?
I might be totally wrong, and I acknowledge that network storage and workload virtualization have been making strides of late, but the only collapse of infrastructure I know of involves fire trucks, ambulances and many, many wounded after an earthquake.
Somehow I doubt that switching between Amazon and Azure is going to be a question of flipping a switch, even in this rosy picture.
More buzzword rubbish. IT isn't Bi-Modal if anything, it's Multi-Modal for a start, and it's really about the services - I.T. has to provide the services demanded by it's users, so if you have a set of users which use an old app, so be it, whist having another set who live at the bleeding edge...
One persons I.T. system is another's archaeology - but if it does the job...
That article seemed to change the subject a couple of times.
But anyway, here's my view on bimodal IT:
Sometimes you can iterate, fail fast and work in an agile way. A large percentage of customer facing IT systems fit this mode. DevOps is the way to go for this.
Sometimes you have to get it right first time, get your requirements right, your design signed off, your Dev and test separated and pretty regimented and deploy once.
For that, Waterfall is the right tool.
Good luck updating a financial reconciliation system using DevOps, for example.
Common sense, innit? Different tools for different jobs.
> you have to offer a single horizontal infrastructure that can be quickly configured to offer the right kind of resource for each one of them when needed.
The way I read this article, it sounds like it's proposing a sort of meta-virtualization: an underlying platform on which the other platforms can run.
I don't see the reason for that. Each of the higher levels like vmware, openstack or containers (e.g. coreOS) can easily have tin added and tin removed. If your vmware cluster is under-utilised you can remove a box and add it to your openstack cluster, or vice versa. Where's the problem?
If it's not proposing running vmware and openstack simultaneously *on the same box*, then what is it proposing? Is it just a matter of how to re-provision boxes to change their function? Any decent shop running at scale will have an automated system build process.
In the mainframe world, funky meta-virtualization stuff is done with nested operating systems. But in the mainframe world you have one big box to carve up; in the data centre world you have lots of small boxes to aggregate.
> Technologies like Software-defined Networking, QoS, advanced storage analytics and monitoring, must be properly implemented to enable such a paradigm…
If "Software-defined Networking" and "QoS" are the solutions, then you probably don't understand the problem.
The article starts with a tongue in cheek rub at bi-modal IT, suggesting it's a buzzword to add to bullshit bingo. And ends with setting off a bingo in the last paragraph with absolutely seriously falling back on software defined networking and storage.
Like two or more articles in one.