I am little old and less outdoorsy now, but back in the day I used to do a fair amount of hiking. The kind of hiking that involves you dehydrating all your food so you don’t have to lug around all those extra kilos as you try to navigate your way through the countryside. After de-camping from the puddle I slept in the night before and start trudging up a hill, one thought always sprang into my mind - pizza. It was always pizza.
Now, I’m not sure about your appetite, but when I need a pizza I don’t rush down the shop and order twelve of them. I did once order more than 30 pizzas when my football team came over to watch a big game, but when it’s just myself and the family, one or two will do nicely.
It has occurred to me (not on a long hike, but while eating a pizza) that NetApp has been asking our customers to buy the data management equivalent of thirty pizzas, when they may have actually needed a lot less. There are good reasons why we would do this - we want to protect your data with RAID-DP and we need to ensure that we have some spare disks in case one of the little blighters fails on you. The problem really isn’t unique to NetApp - if you only need a small amount of data storage then most vendors have a challenge providing this to you efficiently. But this is all about to change at NetApp.
We recently launched our new FAS2500 series entry-level platforms, complete with all the connectivity you need to leverage the benefits of Clustered Data ONTAP. With the launch of Clustered Data ONTAP 8.3 we change the game when it comes to low capacity entry-level configurations.
To explain this, let’s revert back to the pizza analogy.
We have introduced what the engineers are calling Disk Slicing and it will enable us to sell systems with as few as twelve disks, something we really couldn’t do in an active/active system previously. The clever bit is that instead of requiring complete disks for parity, root volumes and spares, we effectively make each disk in the system look like a pizza. We then allocate a slice of each pizza to a “root partition” and the rest of the slices to the “data partition”. We can create parity or spare partitions inside the root partition, along with the root volumes of course.
This has a dramatic effect on the usable capacity for NetApp’s entry-level systems.
For the twelve disk configurations, the usable capacity dramatically increases from 17 per cent to 72 per cent> In a real world example, this means:
That equates to an increase of 436 per cent for the same price point, using Clustered Data ONTAP 8.3.
The good news continues with our All Flash FAS (AFF) configurations. We can utilise the same Disk Slicing feature to enable us to get the best utilisation from smaller AFF configurations, driving the usable capacity to over 70 per cent for 36 and 48 SSD systems. Great news when you consider the price of the larger SDD media.
NetApp hasn’t stopped there. I am assuming most readers are familiar with our FlashPool feature? Effectively it’s our deployment of hybrid storage aggregates where you combine spinning disks and SDD media to meet a range of performance requirements. The good news is that you can now slice your SSD investment into four chunks and allocate them to your aggregates as you require. This can stretch your investment and ensure you target your cache where you need it.
While I have focused on the new entry-level FAS2500 systems, NetApp is supporting a range of platforms for the Disk Slicing functionality –
If any of these solutions fit a challenge you are facing – grab a slice of NetApp.