Confronted with ever increasing amounts of data, IT departments have never been more challenged to find greater capacity and higher resilience in less time and under budget. IT managers are battling to get the most out of their hardware at a time when budget constraints are hitting almost every IT department across the UK.
However, there are ways IT departments can improve and boost the performance of their existing storage hardware by using the right tool for the job. There are four key methods businesses can use and each has its strengths and weaknesses.
Option 1: Adding Solid State Disks to existing enterprise storage
Most traditional storage vendors offer the ability to add Solid State Disks (SSDs) to existing Storage Area Network (SAN) storage. However, while this is technically possible, it comes with limitations on the number of drives that can be deployed due to possible back-end bandwidth limitations. It's also important to stress that traditional storage arrays have been designed with one challenge in mind: getting the most from Hard Disk Drives (HDDs). Adding SSDs into an existing architecture can often cause serious damage to the whole system.
While this option presents itself as a simple and relatively low risk approach to address bottlenecks, due to its use of existing storage management practices, it can prove a costly way of addressing the issue and restrict future growth due to the technical challenges involved.
Option 2: Adding Flash Cards to servers
Opening servers and adding flash cards, such as PCI Express Cards, to the storage design is often thought to be the simplest way to address performance challenges. This approach can be useful for single-application, single-server environments, but adding flash cards to servers is not well suited to enterprise solutions. In fact, adding flash cards to the server can significantly limit the growth and functionality of any end-to-end virtualised solution, due to the constraints found in virtualised environments where tools such as vMotion cannot be deployed as a result of captive storage.
Like option 1, adding flash cards to servers significantly hinders expansion. If growth is required in the long or short term, storage and/or application administrators will need to manually move data between tier 0 and tier 1 and/or utilise a costly data movement tool. This solution can restrict direct data and solution growth, but is perceived to be low risk because only the host configuration is being modified, not the enterprise storage architecture itself.
Adding flash cards to servers is also a high cost approach because of the cost (per/GB) of the cards themselves. Given the budget constraints they are operating under, IT departments may want to think twice about adopting this approach.
Option 3: Utilising All Flash Arrays
In the last year or so, we have seen a plethora of All-Flash Arrays (AFAs) come into the storage market. Whilst AFAs can provide extremely high Input/Output (I/O) rates, they are really designed for one job – high I/O, low latency data delivery. In some cases, such as low latency trading, the cost outlay (often in excess of $100k/TB) can be justified. However, in most cases, projected costs of $100k/TB simply minimise and distort the cost justification for standard enterprise class projects.
Utilising all flash arrays is probably the most unbalanced approach when it comes to building on existing hardware. Adding all flash to existing systems requires extremely high capital and operating expenditure, yet it doesn't deliver the flexibility you might expect. With IT budgets always being squeezed, businesses need to look at their long-term development plans before jumping into the deep-end with a solution that does not necessarily support future growth.
Option 4: Utilising Hybrid storage
By combining flash and HDDs into a single pool, the most appropriate tool can be utilised for the data in hand, be it cache memory or solid state. However, even within hybrid storage architectures, there are different approaches to incorporating flash and controlling data access patterns. The key element here is the ability to move data between different tiers, in real time, without any manual intervention. This should be based on I/O activity rather than workload predictions based on file activity.
In essence, hybrid storage with real time tiering is the only methodology to address the storage architecture challenges presented by new technologies, providing a true balance of cost, growth and risk.
Whether implementing a whole system refresh or adding to an existing system, the trick is to ignore the hype, look at the business needs and make a decision based on cost, risk and anticipated growth.
Gavin McLaughlin, Solutions Development Director at X-IO Storage.
