There is no doubt that the amount of data we have, manage, control and analyse continues to grow ever more rapidly and much of this is unstructured data, documents, images, engineering drawings, data that often needs to be stored in one place and be easily accessible.
However, this presents problems, how do you get all of this data in one place when it’s not just TB’s it 100’s of TB’s and made up of billions of files that need to be accessed quickly, how on earth do you build that kind of capacity and deliver the performance you need?
Like any compute problem, there are two ways to scale things, up by adding more capacity to your existing infrastructure or you can scale out, adding not only more capacity but also more compute.
The other week I heard an excellent episode of the Gestalt IT On-Premise podcast where they posed the question “should all storage be scale out?” (find the episode here) and the answer was basically yes and in a world where we have these ever-growing unstructured data repositories scaling out our NAS services makes perfect sense, delivering not only massive capacity in a single repository, but also taking advantage of scaled-out compute to give us the ability to process the billions of transactions that comes with a huge repository.
So for Episode 50 of the Tech Interviews podcast, it seemed apt to celebrate the big five-oh talking about big data storage.
To discuss this evolution of NAS storage I’m joined by a returning guest, fresh from episode 48 (The heart of the data fabric), Justin Parisi to discuss NetApp’s approach to this challenge, FlexGroups.
We start the episode by discussing what a FlexGroup is and importantly why you may want to use them and why it’s about more than just capacity, as we discuss the performance benefits of spreading our single storage volume across multiple controllers and look at those all important use cases from archives to design and automation.
We explore the importance of simplification, while our need to manage ever-increasing amounts of data continues to grow, the resources available to do it are ever more stretched, so we look at how NetApp has made sure that the complexity of scale-out NAS is hidden away from the user by presenting a simple, intuitive and quick to deploy technology that allows users to have the capacity without the need to rearchitect or relearn their existing solutions.
We wrap up by looking at some of NetApp’s future plans for this technology, including how it may become the standard deployment volume, simplification of migration and other uses such as VMware datastores.
FlexGroups is a really smart technology designed to simply address this ever-growing capacity and performance problem encountered by our traditional approach to file services and if you are looking at scale-out NAS for your file services, it’s a technology well worth reviewing.
For some very informative FlexGroup blogs visit NetApp Newsroom.
There is also a selection of NetApp Technical Documents around the subject, check out TR’s 4557, 4571 and 4616.
You can also hear more from Justin and the Tech ONTAP podcast team discussing FlexGroups here in episode 46.
And finally, you can contact the FlexGroup team via email at firstname.lastname@example.org
If you want to find out more about Justin and the work he does in this area you can check out his excellent website https://whyistheinternetbroken.wordpress.com/ and follow him on Twitter @nfsdudeabides.
Next week It’s the last show of the year as I’m joined by Jason Benedicic and Ruairi Mcbride to discuss the future of datacentre architecture as we talk next-gen technologies.
To catch that show why not subscribe in any of the usual podcast places.
Hope you enjoyed Episode 50 – here’s to the next 50!
Thanks for listening.