Storagebod Rotating Header Image


An Opening is needed

As infrastructure companies like EMC try to move to a more software oriented world; they are having to try different things to try to grab our business. A world where tin is not the differentiator and a world where they are competing head-on with open-source means that they are going to have to take a more open-source type approach. Of course, they will argue that they have been moving this way with some of their products for sometime but these have tended to be outside of their key infrastructure market.

The only way I can see products like ViPR in all it’s forms gaining any kind of penetration will be for EMC to actually open-source it; there is quite a need for a ViPR like product, especially in the arena of storage management but it is far too easy for their competitors to ignore it and subtly block it. So for it to gain any kind of traction; it’ll need open-sourcing.

The same goes for ScaleIO which is competing against a number of open-source products.

But I really get the feeling that EMC are not quite ready for such a radical step; so perhaps the first step will a commercial free-to-use license; none of these mealy mouthed, free-to-use for non-production workloads but a proper you can use this and you can put it into production at your own risk type license. If it breaks and you need support; these are the places you can get support but if it really breaks and you *really* need to to pick up the phone and talk to somone, then you need to pay.

It might that if you want the pretty interface that you need to pay but I’m not sure about that either.

Of course, I’m not just bashing EMC; I still want IBM to take this approach with GPFS; stop messing about, the open-source products are beginning to be good enough for much, certainly outside of some core performance requirements. Ceph for example is really beginning to pick-up some momentum; especially now that RedHat have bought Inktank.

More and more, we are living with infrastructure and infrastructure products that are good enough. The pressure on costs continues for many of us and hence good enough will do; we are expected to deliver against tighter budgets and tight timescales. If you can make it easier for me, by for example allowing my teams to start implementing without a huge upfront price negotiation; the long-term sale will have less friction. If you allow customers to all intents and purposes use your software like open-source; because to be frank, most companies who utilise open-source are not changing the code and could care less whether the source is available; you find that this will play well in the long-term.

The infrastructure market is changing; it becomes more a software play every week. And software is a very different play to infrastructure hardware..


Hype Converges?

In a software-defined data-centre; why are some of the hottest properties, hardware platforms? Nutanix and Simplivity are two such examples that lead to mind; highly converged, sometimes described as hyper-converged servers.

I think that it demonstrates what a mess our data-centres have got into that products such as these have any kind of attraction. Is it the case that we have built in processes that are so slow and inflexible; that a hardware platform that resembles nothing more than a games-console for virtualisation has an attraction.

Surely the value has to be in the software; so have we got so bad at building out data-centres that it makes sense to pay a premium for a hardware platform and there is certainly a large premium for some of them.

Now I don’t doubt that deployment times are quicker but my real concern is why have we got to this situation. It seems that the whole infrastructure deployment model has collapsed under it’s own weight. But is the answer expensive converged hardware platforms?

Perhaps it is time to fix the deployment model and deploy differently because I have a nasty feeling that many of those people who are struggling to deploy their current infrastructure will also struggle to deploy these new hyper-converged servers in a timely manner.

It really doesn’t matter how quickly you can rack, stack and deploy your hypervisor if it takes you weeks to cable it to to talk the outside world or give it an IP address or even a name!

And then the questions will be asked….you couldn’t deploy the old infrastructure in a timely manner; you can’t deploy the new infrastructure in a timely manner even if we pay a premium for it….so perhaps we will give public cloud a go.

Most of problems at present in the data-centre are not technology; they are people and mostly process. And I don’t see any hardware platform fixing these quickly….

Announcement Ennui

Despite my post of approval about IBM’s V7000 announcements; there’s a part of me who wonders who the hell really cares now? The announcements from IBM, HDS, NetApp and the  inevitable EMC announcement later in the year just leave me cold. The storage-array is nearly done as a technology; are we going to see much more in way of genuine innovation?

Bigger and faster is all that is left.

Apart from that, we’ll see incremental improvements in reliability and serviceability. It does seem that the real storage innovation is going to be elsewhere or down in the depths and guts; practically invisible to the end-user and consumer.

So things I expect to see in the traditional array market include a shift from a four-five year refresh cycle for centralised storage arrays to a six-seven year refresh cycle; centralised arrays are getting so large that migration is a very large job and becomes an increasingly large part of an array’s useful life. We will see more arrays offering data-in-place upgrades; replacement of the storage controllers as opposed to the back-end disk.

An Intel-based infrastructure based on common commodity components means that the internal software should be able to run on more generations of any given hardware platform.

We’ll also see more work on alternative to the traditional RAID-constructs; declustered, distributed, micro-RAIDs.

There are going to be more niche devices and some of the traditional centralised storage array market is going to be taken by the AFAs but I am not sure that market is as large as some of the valuations suggest.

AFAs will take off once we have parity pricing with spinning rust but until then they’ll replace a certain amount of tier-1 and find some corner-cases but it is not where the majority of the world’s data is going to sit.

So where is the bulk of the storage market going?

Object storage is a huge market but it is mostly hidden; the margins are wafer-thin when compared to the AFAs and it does not directly replace the traditional centralised array. It is also extremely reliant at the moment on application support. (Yes, I know everyone…I’m a broken record on this!).

The bulk of the storage market is going to be the same as it is now….DAS but in the form of ServerSAN. If this is done right, it will solve a number of problems and remove complexity from the environment. DAS will become a flexible option and no longer just tiny silos of data.

DAS means that I can get my data as close to the compute as possible; I can leverage new technologies in the server such as Diablo Memory Channel Storage but with ServerSAN, I can also scale-out and distribute my data. That East/West connectivity starts to become very important though.

Storage refreshes should be as simple as putting in a new ServerSAN node and evacuating an older node. Anyone who has worked with some of the cluster file-systems will tell you that this has become easy; so much easier than the traditional SAN migration. There is no reason why a ServerSAN migration should not be as simple.

I would hope that we could move away from the artificial LUN construct and just allocate space. The LUN has become increasingly archaic and we should be no longer worrying about queue depths per LUN for example.

There are still challenges; synchronous replication over distance, a key technology for active/active stretched clusters is still a missing technology in most ServerSAN cases. Personally I think that this should move up the stack to the application but developers rarely think about the resilience and other non-functional requirements.

And at some point, someone will come up with a way of integrating ServerSAN and SAN; there’s already discussion on how VSAN might be used with the more traditional arrays.

The storage array is certainly not dead yet and much like the mainframe, it isn’t going away but we are going to see an increasing slowing in the growth of array sales; business appetite for the consumption of storage will continue to grow at a crazy rate but that storage is going to be different.

The challenge for mainstream vendors is how they address the challenge of slowing growth and address the new challenges faced by their customers. How do they address the ‘Internet of Things’? Not with traditional storage arrays…


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.


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.


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.


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.


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.


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.



All The Gear

IBM are a great technology company; they truly are great at technology and so many of the technologies we take for granted can be traced to back to them. And many of today’s implementations still are poorer than the original implementations.

And yet IBM are not the dominant force that they once were; an organisational behemoth, riven with politics and fiefdoms doesn’t always lend itself to agility in the market and often leads to products that are undercooked and have a bit of a ‘soggy bottom’.

I’ve been researching the GSS offering from IBM, GPFS Storage Server; as regular readers of this blog will know, I’m a big fan of GPFS and have a fair amount installed. But don’t think that I’m blinkered to some of the complexities around GPFS; yet it deserves a fair crack of the whip.

There’s a lot to like about GSS; it builds on the solid foundations of GPFS and brings a couple of excellent new features into play.

GPFS Native RAID; also known as declustered RAID is a software implementation of micro-RAID; RAID is done at a block level as opposed to a disk level; this generally means that the cost of rebuilds can be reduced and the time to get back to a protected level can be shortened. As disks continue to get larger, conventional RAID implementations struggle and you can be looking at hours if not days to get back to a protected state.

Disk Hospital; by constantly monitoring the health of the individual disks and collecting metrics for them; the GSS can detect failing disks very early on but there is a dirty secret in the storage world; most disk failures in a storage array are not really failures and could be simply recovered from, a simple power-cycle can be enough or a firmware reflash can be enough to prevent a failure and going into a recovery scenario.

X-IO have been advocating this for a long time; this can reduce maintenance windows and prevent unnecessary rebuilds. It should reduce maintenance costs as well.

Both of these technologies are great and very important to a scalable storage environment.

So why aren’t IBM pushing GSS in general; it’s stuffed full of technology and useful stuff?

The problem is GPFS…GPFS is currently too complicated for many, it’s never going to be a general purpose file system. The licensing model alone precludes that; so if you want to utilise it with a whole bunch of clients, you are going to be rolling your own NFS/SMB 3.0 gateway. Been there, done that…still doing that but it’s not really a sensible option for many.

If IBM really want the GSS to be a success; they need a scaleable and supported NAS gateway in front of it; it needs to be simple to manage. It needs integration with the various virtualisation platforms and they need to simplify the GPFS license model…when I say simplify, I mean get rid of the client license cost.

I want to like product and not just love the technology.

Until then…IBM have got all the gear and no idea….


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.


IT’s choking the life out of me.

I’ve been fairly used to the idea that my PC at home is substantially better than my work one; this has certainly been the case for me for more than a decade. I’m a geek and I spend more than most on my personal technology environment.

However, it is no longer just my home PC; I’ve got better software tools and back-end systems; my home workflow is so much better than my work workflow; it’s not even close. And the integration with my mobile devices, it’s a completely different league altogether. I can edit documents on my iPad, my MBA, my desktop, even my phone and they’ll all sync up and be in the same place for me. My email is a common experience across all devices. My media; it’s just there.

With the only real exception of games; it doesn’t matter which device I’m using to do stuff.

And what is more; it’s not just me; my daughter has the same for her stuff as does my wife. We’ve not had to do anything clever, there’s no clever scripting involved, we just use consumer-level stuff.

Yet our working experience is so much poorer; if my wife wants to work on her stuff for her job, she’s either got to email it to herself or use ‘GoToMyPC’ provided by her employer.

Let’s be honest, for most of us now…our work environment is quite frankly rubbish. It has fallen so far behind consumer IT, it’s sad.

It’s no longer the technology enthusiast who generally has a better environment…it’s almost everyone who has access to IT. And not only that, we pay a lot less for it than the average business.

Our suppliers hide behind a cloak of complexity; I’m beginning to wonder if IT as it is traditionally understood by business is no longer an enabler, it’s just a choke-point.

And yes there are many excuses as to why this is the case; go ahead…make them! I’ve made them myself but I don’t really believe them any more…do you?


So you’ve founded a new storage business; you’ve got a great idea and you want to disrupt the market? Good for you…but you want to maintain the same-old margins as the old crew?

So you build it around commodity hardware; you use the same commodity hardware as I can buy off the shelf; basically the same disks that I can buy off the shelf from PC World or order from my preferred Enterprise tin-shifter.

You tell me that you are lean and mean? You don’t have huge sales overheads, no huge marketing budget and no legacy code to maintain?

You tell me that it’s all about the software but you still want to clothe it in hardware.

And then you tell me it’s cheaper than the stuff that I buy from my current vendor? How much cheaper? 20%, 30%, 40%, 50%??

Then I do the calculations; your cost base and your BoM is much lower and you are actually making more money per terabyte than the big old company that you used to work for?

But hey, I’m still saving money, so that’s okay….

Of course, then I dig a bit more…I want support? Your support organisation is tiny; I do my due diligence,  can you really hit your response times?

But you’ve got a really great feature? How great? I’ve not seen a single vendor come up with a feature that is so awesome and so unique that no-one manages to copy it…few which aren’t in a lab somewhere.

In a race to the bottom; you are still too greedy. You still believe that customers are stupid and will accept being ripped off.

If you were truly disruptive….you’d work out a way of articulating the value of your software without clothing it in hardware. You’d work with me on getting it onto commodity hardware and no I’m not talking about some no-name white-box; you’d work with me on getting it onto my preferred vendor’s kit; be it HP, Dell, Lenovo, Oracle or whoever else…

For hardware issues; I could utilise the economies of scale and the leverage I have with my tin-shifter; you wouldn’t have to set-up a maintenance function or sub-contract it to some third party who will inevitably let us both down.

And for software support; well you could concentrate on those…

You’d help me be truly disruptive…and ultimately we’d both be successful…

2014 – A Look Forward….

As as we come to the end of another year, it is worth looking forward to see what if anything is going to change in the storage world next year because this year has pretty much been a bust as to innovation and radical new products.

So what is going to change?

I get the feeling not a huge amount.

Storage growth is going to continue for the end-users but the vendors are going to continue to experience a plateau of revenues. As end-users, we will expect more for our money but it will be mostly more of the same.

More hype around Software-Defined-Everything will keep the marketeers and the marchitecture specialists well employed for the next twelve months but don’t expect anything radical. The only innovation is going to be around pricing and consumption models as vendors try to maintain margins.

Early conversations this year point to the fact that the vendors really have little idea how to price their products in this space; if your software+commodity-hardware=cost-of-enterprise-array, what is in it for me?  If vendors get their pricing right; this could be very disruptive but at what cost to their own market position?

We shall see more attempts to integrate storage into the whole-stacks and we’ll see more attempts to converge compute, network and storage at hardware and software levels. Most of these will be some kind of Frankenpliance and converged only in shrink-wrap.

Flash will continue to be hyped as the saviour of the data-centre but we’ll still struggle to find real value in the proposition in many places as will many investors. There is a reckoning coming. I think some of the hybrid manufacturers might do better than the All-Flash challengers.

Hopefully however the costs of commodity SSDs will keep coming down and it’ll finally allow everyone to enjoy better performance on their work-laptops!

Shingled Magnetic Recording will allow storage densities to increase and we’ll see larger capacity drives ship but don’t expect them to appear in mainstream arrays soon; the vibration issues and re-write process is going to require some clever software and hardware to fully commercialise these. Still for those of us who are interested in long-term archive disks, this is an area worth watching.

FCoE will continue to be a side-show and FC, like tape, will soldier on happily. NAS will continue to eat away at the block storage market and perhaps 2014 will be the year that Object storage finally takes off.

Five Years On (part 3)

So all the changes referenced in part 2, what do they mean? Are we are at an inflection point?

The answer to the latter question is probably yes but we could be at a number of inflection points both localised vendor inflection points but also industry-wide ones as well. But we’ll probably not know for a couple more years and then with hindsight we can look back and see.

The most dramatic change that we have seen in the past five years is the coming of Flash-based storage devices; this is beginning to change our estates and what we thought was going to become the norm.

Five years ago; we were talking about general purpose, multi-tier arrays; automated tiering and provisioning but all coming together in a single monolithic device. The multi-protocol filer model was going to become the dominant model; this was going to allow us to break down silos in the data centre and to simply the estate.

Arrays were getting bigger as were disks; i/o density was a real problem and generally the slowest part of any system was the back-end storage.

And then SSDs began to happen; I know that flash-based/memory-based arrays have been around for a long time but they were very much specialist and a niche market. But the arrival of the SSD; flash in familar form-factor at a slightly less eye-watering price was a real change-bringer.

EMC and others scrambled to make use of this technology; treat them as a faster disk tier in the existing arrays was the order of the day. Automated Storage Tiering technology was the must have technology for many array manufacturers; few customers could afford to run all of their workloads on an entirely SSD-based infrastructure.

Yet if you talk to the early adopters of SSDs in these arrays; you will soon hear some horror stories; the legacy arrays simply were not architected to make best use of the SSDs in them. And arguably still aren’t; yes, they’ll run faster than your 15k spinning rust tier but you are not getting the full value from them.

I think that all the legacy array manufacturers knew that there were going to be bottle-necks and problems; the different approaches that the vendors take almost points to this and the different approaches taken by a single vendor..from using flash as a cache to utilising it simply as a faster disk…using it as extension of the read cache to using it as both a read and write cache.

Vendors claiming that they had the one true answer….none of them did.

This has enabled a bunch of start-ups to burgeon; where confusion reigns, there is opportunity for disruption. That and the open-sourcing of ZFS has built massive opportunity for smaller start-ups, the cost of entry into the market has dropped. Although if you examine many of the start-ups offerings; they are really  a familiar architecture but aimed at a different price point and market as opposed to the larger storage vendors.

And we have seen a veritable snow-storm of cash both in the form of VC-money but also acquisition as the traditional vendors realise that they simply cannot innovate quickly enough within their own confines.

Whilst all this was going on; there has been an incredible rise in the amount of data that is now being stored and captured. The more traditional architectures struggle; scale-up has it’s limits in many cases and techniques from the HPC market place began to become mainstream. Scale-out architectures had begun to appear; firstly in the HPC market, then into the media space and now with the massive data demands of the traditional enterprises…we see them across the board.

Throw SSDs, Scale-Out together with Virtualisation; you have created a perfect opportunity for all in the storage market to come up with new ways of fleecing providing value to their customers.

How do you get these newly siloed data-stores to work in harmonious and easy to manage way? How do we meet the demands of businesses that are growing ever faster. Of course we invent a new acronym that’s how….’SDS’ or ‘Software Defined Storage’

Funnily enough; the whole SDS movement takes me right back to the beginning; many of my early blogs were focused on the terribleness of ECC as a tool to manage storage. Much of it due to the frustration that it was both truly awful and was trying to do to much.

It needed to be simpler; the administration tools were getting better but the umbrella tools such as ECC just seemed to collapse under their own weight. Getting information out of them was hard work; EMC had teams devoted to writing custom reports for customers because it was so hard to get ECC to report anything useful. There was no real API and it was easier to interrogate that database directly.

But even then it struck me that it should have been simple to code something which sat on top of the various arrays (from all vendors); queried them and pulled back useful information. Most of them already had fully featured CLIs; it should have been not beyond the wit of man to code a layer that sat above the CLIs that took simple operations such as ‘allocate 10x10Gb LUNs to host ‘x’ ‘ and turn them into the appropriate array commands; no matter which array.

I think this is the promise of SDS. I hope the next five years will see the development of this; that we see storage with in a data-centre becoming more standardised from an programmatic point of view.

I have hopes but I’m sure we’ll see many of the vendors trying to push their standard and we’ll probably still be in a world of storage silos and ponds…not a unified Sea of Storage.