Editor's note: This is a 5 part series published by FalconStor. Discussing about legacy storage, they look at whether there is still a future for it in the current market and explored some suggestions of how the industry is and should be changing. The DSA editorial department thinks this is an interesting analysis looking at how the storage industry is changing, for better or for worse.
All opinions expressed below belong to FalconStor.
Is traditional storage dying out?
If you follow IT press, you will have noticed that in recent times the amount of column inches devoted to discussing new storage technologies has increased significantly. We are seeing numerous technologies emerge that are seriously challenging the traditional ways to store enterprise data. Cloud, Hyper-converged Infrastructure and open source software based storage solutions are not just the hype; they are a reality today. The under-current of the IT press and analyst commentary is that these new technologies are so compelling that they are quickly making traditional enterprise disk arrays irrelevant. It seems that the days of the traditional approach are numbered, but is this speculation or fact?
Inflated margins and customer lock in – the dark secret of traditional storage arrays.
Traditional storage vendors have a problem. They acknowledge new technologies and even market them, but the truth is their revenue streams are built on legacy traditional storage arrays, and if they do not keep selling those arrays and the components that go with them, their revenues will decline steeply, ultimately it is not in their interests to allow their existing customers to move away from the traditional storage.
The truth is that the pressure of delivering shareholders profits and meeting revenue commitments is dictating the product strategy of traditional storage vendors. New storage technologies like software defined storage offer flexibility and scalability at lower pricing on commodity components. But advocating this approach to their existing customers would mean traditional storage vendors will drive lower revenue and profits from the customer base. The financial incentive is to keep them locked in on traditional storage for as long as possible.
“Join us or die.”
Traditional arrays are pretty much proprietary. If you want to use your vendor’s replication solution, you MUST replicate to another identical array from that same vendor. If you need to replace a faulty disk, you must use the disk from that vendor. You pay a big premium for this but the actual disk inside is the same disk you can buy from Seagate or western digital for perhaps three percent of the price.
It’s the same with all components, if they fail they must be replaced with components from your branded vendor. Invariably they will place a price premium on these components and outside of the fact that their customers are locked in you have to question how and why these higher prices can be justified.
Nowhere to run – the curse of premium pricing.
If the inflated prices are hard to accept, it is made worse with these traditional vendors insisting that their own service staff perform upgrades and maintenance, again this adds additional costs.
Traditional Storage arrays lock in customers and once locked in there is no choice but to continue paying high premiums.
They may point to high R&D costs, large support functions, and investment in new products to substantiate the high premiums they charge, but those are wearing thin. As an example, many of these technologies are mature and R&D is about maintaining rather than innovating which should not require such large investment. As for support, we are in an age where relying on third parties to support infrastructure is a retrograde step and expensive overhead.
One thing is for sure, progress cannot be halted and end users cannot be fooled. This business model which supports shareholder pressure to hit revenues and profits is preventing end users from moving to smarter, faster more flexible and more cost effective technologies. End users understand this, and will not accept having this forced on them forever. It is this overwhelming truth that is digging the grave for traditional storage offerings.
Exploiting users for their most valuable asset.
The reason why traditional storage vendors have got away with this practice for so long is because storage is about ultimately the home for the most critical business asset – data. The golden role has historically been don’t spare expense on the hardware that stores and protects the most valuable business asset. Business justification for expensive disk has been built on this premise. However, times are changing.
Lock in will be the downfall
Data remains critical, but flexibility in storing and moving data is becoming even more critical. Applications have become virtualised and IT departments are scaling down their datacenters and they develop hybrid cloud strategies. Software Defined technologies have also opened up the entire enterprise datacenter to utilizing commodity hardware at every level. The need for mobility of data now means that the restrictive and proprietary nature of traditional storage is actually having a reverse effect as flexibility is becoming a critical requirement of managing data.
Software Define Storage, Destruction or Liberation?
The time for traditional storage in the datacenter is running out. For Traditional Storage the Grim Reaper is Software Defined Storage. The economics, scalability and flexibility of a software defined approach cannot be ignored.
Monolithic Traditional Storage Arrays that have not been designed to meet the needs of modern software defined computing. The Datacenter is no longer a physical entity that is situated in a single location and as such storage can no longer be a static device tied to a geographic location. Software Defined Storage frees users from these shackles enabling multiple storage devices spread over different locations to be managed and used as one.
Software Defined Storage will at once destroy the traditional approach and liberate IT managers to manage their data in the software defined world.
The Storage industry will not collapse but it is going to change forever.
We have been here before – Lets learn from the History of Servers.
Just a few short years ago the server market was hardware driven, with applications sitting on expensive, discrete physical servers. Server virtualisation changed the dynamic forever. The server industry has been disrupted and altered forever as companies moved to virtualised environments.
The power of the server is now built into the software layer with increases in compute power being achieved by simply adding more commodity resource into the software layer itself.
Companies like Baidu, Facebook and Google quickly embraced this model as it gave them scalability at low cost along with other benefits such as lower management and maintenance overheads.
The server industry has not disappeared BUT it has changed. Server vendors have all changed their offerings to support server virtualization in addition they embraced the hypervisor vendors and partnered with instead of fighting with them.
For the storage vendors they will have no choice but to do the same. The fundamental difference is that many server vendors had been operating on much thinner margins, unlike Storage companies who are still fighting to defend their high margins.
Traditional storage vendors are going to have to take a brave step and understand that to move forward they will need to take a more open approach and understand that a large part of the value in the new storage world will be in generic software that manages all and any storage.
It also means that using lock in as a way to retain customers is about to disappear forever. Retaining customers for storage hardware will become about providing commodity product at attractive prices,
The value will be in the management layer which will open up management of storage hardware from multiple vendors working in unison.
The IT managers will be set free and will be able to manage storage generically as a service rather paying extortionate amounts and spending valuable resource simply managing cumbersome traditional arrays.
The question remains how will the landscape look going forward, where will the Flash vendors sit and how will enterprise storage in the cloud factor into the new storage eco system? These are questions I try to answer in my next article.