Storage infrastructure wakeup call The storage solutions you run today face many challenges, of which the explosive growth in data volumes is just one. In a recent survey of Register readers, we asked what currently stresses your storage systems, and what needs to be done to ensure your internal customers get the services they need. The overriding conclusion was …
You know, when you click the link it says pretty clearly "By Freeform Dynamics, sponsored by X-IO Technologies". And, to be perfectly honest, there's nothing in Tony's article there that is untowards. Furthermore, I am pretty familiar with the Freeform guys, and they don't screw around. They would have collected the data and done proper analysis on it.
"Sponsored by X-IO"? Sure. I buy that's the case. But that won't make what's in the report invalid. Maybe that's the problem, eh? It's hitting a little too close to home?
>article about meeting modern storage challenges
>vmWare and MS products called out by name
>doesn't even mention CEPH, Gluster, Swift, etc (just "Openstack")
Well that was a waste of my time. On a side note, I do have to agree with Mr. Pott, the linked PDF is very obviously shown as sponsored content and I've never seen a linked whitepaper that wasn't. Anywhere.
You do realise that "CEPH, Gluster, Swift, etc" are, for all intents and purposes "edge cases". A few hyperscale adoptions that boost the number of nodes deployed, but the number of total organizations adopting them are pretty damend small.
Furthermore, in the real world, "CEPH, Gluster, Swift, etc" do basically fall under "Openstack" for most companies. They don't think about the individual storage technologies in play, and they certainly don't think about those storage technologies outside of Openstack itself. (With the possible exception of CEPH, and even then only in some very rare cases.)
Like it or not, "CEPH, Gluster, Swift, etc" are not out there "solving modern storage challenges." They are - maybe - the technologies of tomorrow. But the ecosystem has to evolve a whole lot before any of those technologies become something mainstream businesses can consider...and they'll require a major commercial partner that largely directs and controls them (a-la RedHat.)
I am afraid that Openstack remains "early adopters only" for now, and all of it's attendant technologies sit there too. That's changing, but not enough to be of material significance for this refresh cycle.
@thegreatsatan Good to see people looking closely at the origin of content and the funding behind it, but this really is straight up research and analysis based on Reg reader input. If there is something about it that you think is vendor biased, then you’ll have to be more specific. If it’s just a general objection to the concept of sponsorship as a funding model, then fair enough, but that horse bolted a long time ago.
@AC With respect to emerging OSS / scale-out options, this is something we have been looking at very closely at Freeform Dynamics recently. Our conclusion is the same as Trevor's in terms of readiness for prime time. In most cases you still have to be willing to get your hands very dirty with design, config and integration – fine if you are large scale service provider implementing such options as part of your core business, or a corporate IT department with an obvious need and resources to spare, but not really mainstream at the moment. If you go sniffing around the project sites and discussion groups, you’ll see that a lot of features many enterprises would consider to be ‘core’ are still designated as ‘work in progress’ – it’s all out in the open if you look. Having said that, players like Fujitsu with its Ceph-based scale-out offerings, and various other big vendors (usual suspects) with their efforts around producing enterprise-friendly incarnations of Openstack, for example, are helping things along in mainstream, so it’s all moving in the right direction.
In the meantime, the key message from the research is that just because your storage environment has served you well over the last 10 years doesn’t mean you are sitting pretty to deal with the needs of the next 10. Requirements are changing, and this mandates different approaches and different technologies. Tony Lock’s article was an attempt to walk through some of the specifics behind this as a reminder to some, and a ‘wakeup call’ to others. The full research report linked to in the piece goes into a lot more detail.
Cheers, Dale (Freeform Dynamics)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
