Storagebod Rotating Header Image

October, 2014:

Fujitsu Storage – With Tentacles..

Oct 31st, 2014
by Martin Glassborow.
No comments yet

So Fujitsu have announced the ETERNUS CD10000; their latest storage product designed to meet the demands of the hyperscale and the explosion in data growth and it’s based on….Ceph.

It seems that Ceph is quickly becoming the go-to scale-out and unified system for those companies who don’t already have an in-house file-system to work-on. Redhat’s acquisition of Ink-Tank has steadied that ship with regards to commercial support.

And it is hard to see why anyone would go to Fujitsu for a Ceph cluster; especially considering some of the caveats that Fujitsu put on it’s deployment. The CD10000 will scale to 224 nodes; that’s a lot of server to put on the floor just to support storage workloads and yet Fujitsu were very wary about allowing you to run workloads on the storage nodes despite the fact that the core operating system is Centos.

CephFS is an option with the CD10000 but the Ceph website explicitly says that this is not ready for production workloads; even with the latest release .87 Giant. Yes, you read that right; Ceph is not yet a v1.0 release; now that in itself will scare off a number of potential clients.

It’s a brave decision of Fujitsu to base a major new product on Ceph; it’s still very early days for Ceph in the production mainstream. But with large chunks of IT industry betting on OpenStack and Ceph’s close (but not core) relationship with OpenStack, it’s kind of understandable.

Personally, I think it’s a bit early and the caveats around the Eternus CD10000 deployment is limiting currently; I’d wait for the next release or so before deploying.

Done

Oct 30th, 2014
by Martin Glassborow.
No comments yet

Could VMAX3 possibly be the last incarnation of the Symmetrix that ships?

As an Enterprise Array, it feels done; there is little left to do, arguably this has been the case for some time but the missing feature for VMAX had always been ease of use and simplicity. The little foibles such as the Rule of 17, Hypers, Metas, BCVs vs Clones all added to the mystique/complexity and led to many storage admins believing that we were some kind of special priesthood.

The latest version of VMAX and the rebrand of the Enginuity into HyperMax removes much of this and it finally feels like a modern array…as easy to configure and run as any array from their competitors.

And with this ease of use; it feels like the VMAX is done as an Enterprise Array…there is little more to add. As block array, it is feature complete.

The new NAS functionality will need building upon but apart from this…it’s done.

So this leaves EMC with VNX and VMAX; two products that are very close in features and functionality; one that is cheap and one that is still expensive. So VMAX’s only key differentiator is cost…a Stellar Artois of the storage world.

I can’t help but feel that VNX should have a relatively short future but perhaps EMC will continue to gouge the market with the eye-watering costs that VMAX still attracts. A few years a go; I thought the Clariion team might win out over the Symm team, now I tend to believe that eventually the Symm will win out.

But as it stands, VMAX3 is the best enterprise array that EMC have shipped but arguably it should be the last enterprise array that they ship. The next VMAX version should just be software running on either your hardware or perhaps a common commodity platform that EMC ship with the option of running the storage personality of choice. And at that point; it will become increasingly hard to justify the extra costs that the ‘Enterprise’ array attracts.

This model is radically different to the way they sell today…so moving them into a group with the BURA folks makes sense; these folks are used to selling software and understand that is a different model..well some of them do.

EMC continue to try to re-shape themselves and are desperately trying to change their image; I can see a lot of pain for them over the next few years especially as they move out of the Tucci era.

Could they fail?

Absolutely but we live a world where it is conceivable that anyone of the big IT vendors could fail in the next five years. I don’t think I remember a time when they all looked so vulnerable but as their traditional products move to a state of ‘doneness’; they are all thrashing around looking for the next thing.

And hopefully they won’t get away with simply rebranding the old as new…but they will continue to try.

 

Scrapheap Challenge

Oct 28th, 2014
by Martin Glassborow.
4 comments

On the way to ‘Powering the Cloud’ with Greg Ferro and Chris Evans, we got to discussing Greg’s book White Box Networking and whether there could be a whole series of books discussing White Box storage, virtualisation, servers etc and how to build a complete White Box environment.

This lead me to thinking about how you would build an entire environment and how cheap it would be if you simply used eBay as your supplier/reseller.  If you start looking round eBay, it is crazy how far you can make your money go; dual processor HP G7s with 24Gb for less than £1000.; 40 port 10 GbE switch for £1500; 10 GbE cards down to £60.  Throw in a Supermicro 36 drive storage chassis and build a hefty storage device utilising that; you can build a substantial environment for less than £10,000 without even trying.

I wonder how far you could go in building the necessary infrastructure for a start-up with very few compromises. And whether you can completely avoid going into the cloud at all?  The thing that itsstill going to hurt is the external network connectivity to the rest of the world.

But instead of ‘White Box’…perhaps it’s time for junk-box infrastructure. I don’t think it’d be any worse than quite a few existing corporate infrastructures and would probably be more up-to-date than many.

What you could build?

 

Take Provisions

Oct 23rd, 2014
by Martin Glassborow.
No comments yet

Provisioning and utilisation seem to be a perennial subject; there still seems to be a massive problem with large Enterprises who appear to have far too much storage for the amount of data that they store. There are many reasons for this and there is certainly fault on both the customer and vendor side.

Firstly, as a customer; you should expect to be able to use all the capacity that you purchase and your internal pricing models should reflect this. Obviously, there is always overhead for data-protection and hot-spares but once this is taken into account, all that capacity should be useable. You need to define what useable capacity means to you.

There should not be a performance hit for utilising all useable capacity. If the vendor states that best practise is only to use 70% of useable capacity; that needs to reflected in your TCO calcuations to give you a true cost of storage.

Secondly, as a customer; you need to ensure that your processes and operational procedures are based around provisioning the right amount of storage and not over-provisioning. Allocating storage for a rainy day is not a great idea; thin provisioning can help with this but it is not a silver bullet.

Thirdly, as a vendor; you need to be up front about the useable capacity; if you can only realistically use 70%, you need to factor this into your pricing models and show the customer exactly what they are getting for their money. Be open and honest. If you want to show a price per IOP, a price per gigabyte or some other measure; that is fine. If you want to show a price based on your assumed dedupe ration; be prepared to put your money where your mouth is.

Fourthly, as a vendor; look at ways of changing how storage is allocated and presented. It is time for us to move away from LUNs and other such archiac notions; provisioning needs to be efficient and simple. And we also need the ability to deallocate as easily as we allocate. This has often been problematic. Obviously this is not just a storage problem; how many companies are spinning up VMs and then not clearing them down properly? But make it easier across the board.

Fifthly, as vendors and users; we need to look at data-mobility. Too often, the reason that an array is under-utilised is because you have reserved capacity for application growth because it is simply ‘too hard’ to move an application’s data once it is in place. This is also a reason why many customers are very wary about thin-provisioning; the rainy day scenario again.

However, large arrays bring their own issues; from change management to refresh cycles. Smaller might be better for many but unless data can be easily moved; there is a tendency to buy arrays that are large and reserving capacity for growth.