Driven by the evolving needs of a modern, mobile, social and data-driven workforce, businesses are set to see a step change in the way they allocate IT budgets.
According to a new IDC report, it is predicted that over the next five years more than 95 per cent of all business IT expenditure will contribute towards “3rd Platform" computing infrastructure – incorporating technologies such as cloud, mobile, social and big data.
One common thread is prevalent across the third platform: the need for vast, accessible and economic solutions to host huge banks of data. The problem for businesses and data centre operators is that traditional storage solutions are simply not suited to handling the increased complexity of processes required for the current and emerging business world.
IDC explains that this increased need for vast expanses of data storage is a key factor behind the changing IT spending habits of businesses. Traditional storage facilities, made up of stacks of hard-disk drives (HDD), were once the be-and-end all of data storage. Even today, they are serviceable for environments that require a low dollar-per-gigabyte cost for capacity and need relatively few inputs/outputs per second (IOPS). However, using this same approach in today’s data intensive world simply doesn’t work. Why? That’s where the economics of flash data storage come in to play.
The argument against flash always starts with its price point. Flash, with its higher cost per gigabyte association has often been priced out of the datacentre, and thus out of the enterprise storage solution equation. But developments in technology, as well as an evolution of workforce habits, have changed things. IDC has spoken at length about the potential of flash to dominate spinning disk in performance intensive storage environments. To this end, we’re already seeing flash storage transforming the world’s datacentres.
Put very simply, flash storage devices enable higher performance access to data, which is absolutely critical to today’s fast moving businesses. An all-flash array, that is a solid state storage disk system containing multiple flash memory drives instead of spinning hard disk drives, can transfer data to and from solid state drives much faster than hard disk drives.
Where traditional HDD storage often misses the mark is ensuring consistently low latency and through unnecessary delays experienced when customers try to dip in and out to access their data. Flash storage allows a far more seamless way for organisations to access their data when and where they need it - without that extra latency or delay.
Flash-based solid state storage devices also require much less power for running and cooling than HDDs, especially once IT managers run de-duplication. With power in the data centre usually being placed in the top three IT operating costs - alongside HR overheads and datacentre space - any reduction in energy costs will generate significant savings to the business.
All-flash arrays have a higher performance throughput, which means datacentres and businesses need fewer drives to manage the same workloads. Additionally, as workload capacities demand higher throughput rates (measured in inputs/outputs per second or IOPS), many enterprises are finding their workloads are perfectly positioned for flash's cost-to-performance ratio.
But what does this all mean for businesses? In simple economics, combining all the benefits of flash brings manifold returns, dispelling the myth that flash is too expensive for the enterprise. Deploying flash within storage arrays built to take advantage of flash based media will result in lower storage administration. Revolutionary internal architecture completely eliminates complex set-up and tuning steps, while inherently delivering maximum performance. Given that a single tier of flash based media will be deployed there is no more management of storage tiers and the ‘chasing of performance hotspots’.
Clearly then flash is the only storage solution that can cope with the demands of the third platform. Yet until recently, it has not been available at a reasonable enough price point for many, as the dollar-per-gigabyte cost has been a large barrier to entry for businesses. Though the absolute cost of flash storage has been coming down significantly over the last few years, making the relative cost per TB of storage more attractive than ever in a flash vs. traditional array, businesses have been unable to justify the cost of flash – so have missed out on the multitude of secondary benefits.
In order to make flash appealing to these businesses, it is important to highlight and understand how these secondary benefits - such as lower device count, lower energy consumption and reduced floor space requirements - can actually equate to lower overall operating costs, bringing down the TCO for businesses.
In economic terms, if one makes the conservative assumption that flash can achieve a data reduction ratio of 2:1 for secondary stage workloads, along with reducing number of devices by a conservative 20 per cent, a flash-based configuration would achieve 40-80 per cent lower TCO.
In real terms, a flash-based medium available at roughly $0.40 per gigabyte for raw capacity (by 2018) would cost roughly the same as a storage solution built out of performance optimised HDDs priced at $0.183 per gigabyte. To compare a flash-based configuration with a storage solution built entirely out of capacity-optimised HDDs, we would have to assume that we’d need 4 times the number of devices to meet IOPS requirements, a number that significantly raises not only the acquisition cost but also the TCO of the solution.
This is just a quick analysis of course, but we can already start to formulate a view of just how viable storage is in today’s third platform environment. The storage product would have the secondary economic benefits and cost $0.40 per gigabyte, by 2018. It would deliver performance (latencies, IOPS, and throughput) that is consistently better than the best case that performance-optimised HDDs can offer today. And since this would be targeted at secondary storage environments, it would be built for scalability into the tens or hundreds of petabytes for application workloads that require high ingest rates, exhibit low change rates, and need high read rates with some level of intermixed random writes.
Targeted as it would be at massively scalable environments, IDC suggests that an appropriate name for this new category would be "Big Data Flash." This offers a significant shift in economics; that, more often than not, is the biggest factor in the process of adopting new technology.
The creation of the new category of Big Data Flash is a welcome development in an industry that is struggling with legacy HDD technologies for secondary storage applications in the third platform computing era.
The performance and cost characteristics of Big Data Flash enable not only new types of secondary storage applications but possibly also new types of primary storage applications that meet the needs of the modern workforce.
Marcos Burnett, Sales Director for Northern Europe, SanDisk.