Big Ideas

So EMC have finally productised Nile and given it the wonderful name of ‘Elastic Cloud Storage’; there is much to like about it and much I have been asking for…but before I talk about what I like about it, I’ll point out one thing…

Not Stretchy

It’s not very Elastic, well not when compared to the Public Cloud Offerings unless there is a very complicated finance model behind it and even then it might not be that Elastic. One of the things that people really like about Public Cloud Storage is that they pay for what they use and if their consumption goes down….then their costs go down.

Now EMC can probably come up with a monthly charge based on how much you are using; they certainly can do capacity on demand. And they might be able to do something with leasing to allow downscaling as well at a financial level but what they can’t easily do is take storage away on demand. So that 5 petabytes will be on premise and using space; it will also need maintaining even if it spins down to save power.

Currently EMC are stating 9%-28% lower TCO over Public Cloud…it needs to be. Also that is today; Google and Amazon are fighting a price-war, can EMC play in that space and react quickly enough? They claim that they are cheaper after the last round of price cutting but after the next?

So it’s not as Elastic as Public Cloud and this might matter…unless they are relying on the fact that storage demands never seem to go away.

Commodity

I can’t remember when I started writing about commodity storage and the convergence between storage and servers. Be it roll-your-own or when vendors were going to start doing something very similar; ZFS really sparked a movement who looked at storage and thought why do we need big vendors like EMC, NetApp, HDS and HP for example.

Yet there was always the thorny issue of support and for many of us; it was a bridge too far. In fact, it actually started to look more expensive than buying a supported product..and we quite liked sleeping at night.

But there were some interesting chassis out there that really started to catch our eyes and even our traditional server vendors were shipping interesting boxes. It was awfully tempting.

And so I kept nagging the traditional vendors…

Many didn’t want to play or were caught up in their traditional business. Some didn’t realise that this was something that they could do and some still don’t.

Acquisition

The one company who had the most to loose from a movement to commodity storage was EMC; really, this could be very bad news. There’s enough ‘hate’ in the market for a commodity movement to get some real traction. So they bought a company that could allow commoditisation of storage at scale; I think at least some of us thought that would be the end of that. Or it would disappear down a rabbit hole to resurface as an overpriced product.

And the initial indications were that it wasn’t going to disappear but it was going to be stupidly expensive.

Also getting EMC to talk sensibly about Scale-IO was a real struggle but the indication is that it was a good but expensive product.

Today

So what EMC have announced at EMC-World is kind of surprising in that it looks like that they may well be willing to rip the guts out of their own market. We can argue about the pricing and the TCO model but it looks a good start; street prices and list prices have a very loose relationship. The four year TCO they are quoting needs to drop by a bit to be really interesting.

But the packaging and the option to deploy on your own hardware; although this is going to be from a carefully controlled catalogue I guess; is a real change from EMC. But you will also notice that EMC have got into the server-game; a shot across the bows of the converged players?

And don’t just expect this to be a content dump; Scale-IO can do serious I/O if you deploy SSDs.

Tomorrow

My biggest problem with Scale-IO is that it breaks EMC; breaks them in a good way but it’s a completely different sales model. For large storage consumers, an Enterprise License Agreement with all you can eat and deploying onto your chosen commodity platform is going to be very attractive. Now the ELA might be a big-sum but as a per terabyte cost; it might not be so big and the more you use; the cheaper it gets.

And Old EMC might struggle a bit with that. They’ll probably try to sell you a VMAX to sit behind your ViPR nodes.

Competitors?

RedHat have an opportunity now with Ceph; especially amongst those who hate EMC for being EMC. IBM could do something with GPFS. HP have a variety of products.

There are certainly smaller competitors as well.

And then there’s VMware with VSAN; which I still don’t understand!

There’s an opportunity here for a number of people…they need to grasp it and compete. This isn’t going to go away any more.

So VSAN is finally here in a released form; on paper, it sure looks impressive but it’s not for me.

I spend an awful lot of time looking at Scale-Out Storage systems; looking at ways to do them faster, cheaper and better. And although I welcome VMware and VSAN to the party; I think that their product falls some-way from the mark but I don’t think that I’m really the target market; it’s not really ready or appropriate for Media and Entertainment or anyone interested in HyperScale.

But even so I’ve got thoughts that I’d like to share.

So VSAN is better because it runs in the VMware kernel? This seems logical but this has tied VSAN to VMware in a way that some of the competing products are not; if I want to run a Gluster Cluster which encompasses not just VMware but also XEN, bare-metal and anything else, I could. And there might be some excellent reasons why I would want to do so, I’d transcode on bare-metal machines for example but might present out on VM-ed application servers. Of course, it is not only Media and Entertainment who have such requirements; there are plenty of other places where heavy lifting would be better done on the bare-metal.

I think that VMware need to be much more open about allowing third party access to the kernel interfaces; they should allow more pluggable options; so I could run GPFS, ScaleIO, Gluster, Stornext within the VMWare kernel.

VSAN limits itself by tying itself so closely to the VMware stack; it’s scalability is limited by the current cluster size. Now there are plenty good architectural reasons for doing so but most of these are enforced by a VMware-only mindset.

But why limit to only 35 disks per server? An HP ProLiant SL4540 takes 60 disks and there are SuperMicro chassis that take 72 disks. Increasing the spindle count not only increases the maximum capacity but the RAW IOps of the solution. Of course, there might be some saturation issues with regards to the inter-server communication.

Yet, I do think it is interesting how the converged IT stacks are progressing; the differences in approach; VMware itself is pretty much a converged stack now but it is a software converged stack; VCE and Nutanix converge onto hardware as well. And yes, VMware is currently the core of all of this.

I actually prefer the VMware-only approach in many ways as I think I could scale computer and storage separately within some boundaries; I’m not sure what the impact of having unbalanced clusters will be on VSAN? Whether it would make sense to have some Big Flipping Dense VSAN appliances rather than distributing the storage equally across the nodes?

But VSAN is certainly welcome in the market; it certainly validates the approaches being taken by a number of other companies…I just wish it were more flexible and open.