The Future of Hyper-converged Infrastructure
Storage continues to be a volatile segment of IT. Hot areas trending in the news this month include NVMe over Fibre Channel, which is being hyped heavily now that the Broadcom acquisition of Brocade is a done deal. Another hot segment is the hyper-converged space, complimented by activity in software-defined storage from several vendors.
Flash is now running ahead of enterprise hard drives in the market, contributing to foundry changeovers to 3D NAND to temporarily put upward pressure on SSD pricing. High-performance storage solutions built on COTS platforms have been announced, too, which will create more pressure to reduce appliance prices.
Let’s cover these topics and more in detail:
- NVMe over Fibre-Channel is in full hype mode right now. This solution is a major step away from traditional FC insofar as it no longer encapsulates the SCSI block-IO protocol. Instead, it uses a now-standard direct-memory access approach to reduce overhead and speed up performance significantly.
Webster defines the word Virtual as “Very close to being something without actually being it.” By that definition, a Virtual SSD is very close to being an SSD without actually being an SSD. From a more practical viewpoint, a Virtual SSD provides the benefits of an SSD, performance, without the downside of an SSD, price and limited capacity.
A virtual SSD blends 2 or more classes of storage media into a single volume with the performance characteristics of the fastest storage media and the capacity of the combined media. The flexibility of a virtual SSD can be tailored to meet a wide variety of needs. Virtual SSDs are:
- Cost effective because you can blend SSDs with cost effective hard drives
- Uncompromising because you can blend high performing NVMe drives with capacity SAS/SATA SSDs for an all flash virtual SSD
- High value because they offer both best in class cost/IO metrics as well as cost/GB
If you could have all flash performance at 80% of the cost of an all-flash solution, would you be interested? From a financial point of view, it is hard to argue.
It is pretty well documented that for most applications, less than 5-10% of the data across a volume is active at any point in time. Given this, it does not make sense to pay all flash prices to store your inactive data. Granted, there are applications that are active across the entire LBA range of a volume, or that cannot tolerate any unpredictable latency. For these applications, by all means, an all-flash solution is the right choice.
For the majority of us, a hybrid solution is the optimal choice. Keep in mind that not all hybrid solutions are created equal. For optimal performance, your hybrid solution should have the following characteristics:
- File pinning because there are applications for which you want to guarantee full flash performance for all IO
- Real time data promotion to performance storage because the solution should react to dynamic requirements, not yesterday’s activity
- Full automation because your time is too valuable to constantly manage storage
- Unrestricted fast tier capacities because you need the ability to size your storage to your unique needs
- Visibility into your storage behavior because you need information to guarantee your storage is optimized
ATTEND THE WEBINAR June 8th at 10:00 AM PDT
Register here: http://goo.gl/LzsgyI
Exponential growth of data from machines and users demands we scale storage infrastructure in real-time, but are you scaling intelligently? Are you optimizing effectively? Have you defined your strategy, your approach?
Great technology is linked to strong partnerships. Each partner brings his own unique expertise and value to the relationship, making the collective stronger than the singular. In that spirit, Enmotus is proud to partner with Intel and join them in the Intel Storage Builders program. Intel Storage Builders provides an open, collaborative environment for true storage innovation and advancement. The program brings together forward-looking companies, a broad Intel ecosystem, key foundational technologies, and strategic resources to empower our members to move storage forward.
When most people think of hybrid storage, they think about flash mixed with HDD, but you can also have an all flash hybrid solution as well. Hybrid All Flash makes sound business sense. It makes sense for the same reasons traditional hybrid solutions make sense. Hybrid all flash solutions provide the performance of high performance flash combined with the economics and capacity points of more economical flash storage.
IDC Corporation recently published a survey stating that 70% of big enterprises across Asia Pacific have or plan to deploy All Flash Arrays (AFA). This shouldn’t be surprising as enterprises have the budgets to deploy AFAs, which allows them to utilize these arrays to achieve operational efficiency, speed application performance and expand their businesses with sophisticated real time analytics and decision support systems. For those organizations that require all flash arrays, a hybrid all flash array will still provide the application performance but it could also improve operational efficiency even more.
Back in college humanities we learned about the Inconsistent Triad. Without getting into a religious discussion, suffice it to say that the Inconsistent Triad consisted of 3 statements. Any of the 2 statements taken together were true, but the third always negated the other 2. I bring this up because I believe we have an Inconsistent Triad in the storage business. The 3 components of the Storage Triad are Performance, Capacity and Cost (or affordability). It is hard to get all 3!
If you desire performance and high capacity, you need to sacrifice affordability
If you desire capacity and affordability, you need to sacrifice performance.
If you desire performance at an affordable price point, you need to sacrifice capacity
This is where hybrid storage comes into play – Hybrid storage balances Performance, Capacity and Cost.
The drop in price for solid state drives continues to fuel the adoption rate of this storage media. While the debate continues about if and when solid state will replace hard drives, it is clear that in the majority of applications, tiering solid state storage with HDDs into a hybrid volume makes economic sense today.
The true cost of storage takes into account both the acquisition costs of hardware as well as ongoing operating costs, with power and cooling accounting for a significant portion of the operating costs. Over a 5 year period, a scale out storage architecture which incorporates tiered hybrid storage saves significant costs over both a pure HDD solution as well as an all flash solution.
One of the most common misconceptions about tiering that I hear is that it is “Difficult”. Other frequent comments include statements such as “Tiering requires IT managers to configure the tiers for a specific workload”, and “For flexibility and growth, caching is the answer”. These misconceptions are based on most people’s knowledge of tiering, which I will call “Macro” Tiering and is most prevalent in enterprise storage arrays.
The reality is that today’s solutions make tiering very simple, and in most cases are the better option for applications that would normally consider caching. With ever increasing growth in server storage, as well as scale out storage, we have seen tiering migrate from high-end external storage into server-based storage. This movement into servers necessitates changes in tiering algorithms and how they behave, as moving large files based on yesterdays activity or even last weeks activity is both inefficient and does not respond to the dynamic environment of most workloads.