Hyperscale Computing Paradigms Hit Mainstream

By William Tan, Head of Enterprise Solutions, Dell Malaysia

It was nearly a decade ago when a new breed of customer – Internet companies building giant datacenter capacity – found themselves in need of a new type of server to support their massive scale.

This trend certainly didn’t go unnoticed. In fact, in 2007, a few entrepreneurial-minded engineers at Dell saw this opportunity and created a new business targeted at addressing these specific needs.  Fast forward a few years, and a new server market segment was created with new players continuing to enter the ring today.

But web giants like Google and Amazon aren’t the only ones who need simple, scalable IT.  Businesses around the globe have been paying attention to the flexibility and efficiency these web companies have achieved with their IT strategies, but are curious how to apply the hyperscale design principles to their more mainstream datacenter needs.

William Tan

In addition, as people change jobs – sometimes moving out of large hyperscale IT operations and into more mainstream enterprises – they often bring the hyperscale values with them to make their IT environments more efficient, flexible and simple. It’s not just about the lowest possible cost for a system, but also having the right robust IT architecture that improves total cost of ownership and maximizes the value they provide to their constituents.

So while it’s true there are still sharp contrasts between hyperscale and mainstream enterprise computing, leading enterprises are starting to adopt the guiding principles of hyperscale.

A closer look

In order to better understand this evolution, it’s important to understand the server market dynamics and how they have changed over the past few years.  Let’s simplify things and say there are four primary market segments that have emerged as major forces in the server market today:  First, there are the largest of the large global Internet companies that we’ve already discussed, then there are the companies that aren’t quite hyperscale, but still massive in scale such as Web Tech and HPC, next up is traditional large enterprises, followed by small and medium businesses.

These server market segments all have their own unique challenges, needs and workload requirements that need to be addressed when bringing new solutions to market. But that’s not to say best practices shouldn’t be shared or cross-pollination shouldn’t happen.

In the case of hyperscale, we witnessed first-hand when working with the massive Internet companies that they needed the ability to keep their IT architecture common, but tweak server, storage and networking for certain workloads like analytics or web serving. It simply didn’t make sense from a CapEx, OpEx or efficiency standpoint for them to build different architectures for different workloads. This same principle applies to the other market segments as well.

Take large enterprises and big data as an example. Business units will increasingly demand capabilities that involve real-time analytics and greater intelligence in their IT operations.  The marketing department may want to deliver sales messages to mobile phones based on a customer’s purchasing history, social media activity, location and other sensor data.  But that same marketing department will also want to continue running its conventional CRM system and dozens of other applications on the same infrastructure.  The bottom line is datacenters must have the flexibility to meet all these varying needs, and also to handle usage spikes by shifting resource utilization rather than by maintaining costly excess capacity that can go unused for significant amounts of time.

New Converged Architectures

When you’re working with datacenters that are comprised of literally, hundreds of thousands of servers there’s a textbook of lessons waiting to be learned and that’s exactly what happened. 2015 marks the year that IT leaders can gain the advantages of hyperscale technologies that cloud providers have achieved.

Case in point, new converged architectures are available that provide – for the first time – a common, scalable platform that easily adapts to the ever-shifting business and technology landscape. By leveraging such a building-block concept derived from the large hyperscale IT operations, organizations can better manage, scale and tailor infrastructure to meet business needs as they change over time.

While technology to converge server, storage and networking is already here, these new converged architectures will revolutionize how modern enterprises consume and manage IT in 2015.

Tweet this byline using this handle: @DellBizSA