Not long ago the primary issues plaguing storage were performance and cost. Those issues have been tackled with hardware innovations, including flash, multicore, and 10GB Ethernet. Now the obstacle is data management, which is still abstracted in LUNs, volumes and other legacy concepts. Since the business thinks in virtual machines (VMs), the natural solution is to manage storage in VMs. The resulting visibility, automation and analytics unlocks the full capabilities of the hardware.
Consider tablets—the hardware technology was available, but it was the simplicity of management (iPad/iOS) that transformed the market. In 2015, storage software will spark the beginnings of a similar transformation.
Virtual environments may be commonplace in the enterprise data centre but the growth in adoption of virtualisation has thrown some of the issues surrounding storage in virtualised environments into stark relief. The fact is, traditional storage is not the best fit for virtual environments. It was designed decades before virtualisation existed and is ill-equipped to support it. It is hard to manage and requires specialists steeped in LUNs, volumes, RAID, etc.
As a result, storage has become a black box, expensive and hard to manage. But the silos of IT are being broken down, and so in 2015, IT generalists – virtualisation admins, system engineers and software developers – will start to take ownership of their server, virtualisation AND storage requirements. In turn, that elevates the storage admin to working on projects with more strategic value than carving up LUNs and volumes.
Another concern we have frequently heard about the last year is the lack of a strong economic case for large-scale deployments and the fact that there will continue to be a challenge of adapting enterprise applications to the cloud, forcing companies to explore private cloud. Many of the capabilities of public cloud will become available in the private cloud—allowing companies to realize scale, security, automation, data protection, and ultimately cost savings. Service providers will quickly grow to service the private clouds of organisations that don’t want to run their own infrastructure.
Next year we will start to see some significant changes in technology. Since 2010 OpenStack has been a point of discussion, but in 2015 it’s become very real. IBM, HP, Cisco, Red Hat, Oracle and VMware all support OpenStack, prompting many companies to start looking at OpenStack seriously. The momentum will start in Software Development (specifically DevOps) where the flexibility/agility of OpenStack is most compelling. Once value is realised it will naturally spread to other parts of the enterprise.
Looking back on 2014, we have seen many people turn to the latest hype – flash. Flash is fast, delivers low latency and can handle random I/Os. But it’s also expensive. Flash storage costs between five and ten times as much as hard disk storage. In addition, flash-only arrays still require separate low-cost disk-based storage systems to store snapshots, replicas, infrequently accessed data and the data of less I/O intensive applications.
In 2015, customers will realise that buying basic flash-based storage doesn’t provide a lot of value, and it doesn’t address the root cause of data centre pain. Why? Because flash performance without baked-in intelligence will be table stakes in 2015—companies will see that driving an orange Ferrari while wearing a blindfold is cool for a short time, but not very bright in the end.
So, they’ll demand total visibility (GPS in our metaphor) into individual VMs; and any flash provider that forces companies to continue to think in LUNs, volumes, etc. will quickly lose its shine. While the case for all-flash arrays in a virtualised environment is hard to sustain, the benefits of smart storage in a virtualised environment are so compelling that, in a few years, people will wonder how they ever got along with it.
We will start to see service providers turn to storage as a competitive advantage. Cloud service providers compete on speed and reliability. In the past that has put the focus on their network. But smart providers have realised that storage can have greater amplification of their performance. They’ll turn to storage that removes guesswork; allowing for precise pricing, quality of service guarantees and happier customers. Storage providers that build upon these sources of differentiation will be best positioned to capture the private cloud opportunity.
We have seen some innovative solutions developed in past year tackling the problems of VDI, troubleshooting errors, and platforms that offer the scalability and visibility needed for high VM workloads. But whatever happens 2015 looks like a very exciting year for the storage industry as it’s due for a shake up.