Primary tab navigation

The Evolution of Hardware and What It Means for Big Data

Imagine being able to use the weather forecast to know which of your promotions will be most effective on any given day. Or using visitor data on your Web site and social media feeds to get a complete, real-time picture of your customers’ likes and dislikes.

These projects that amass and analyze vast volumes of data, or “big data,” as we’ve come to know it, are materializing within midsize firms. Once the sole domain of large enterprises, big data is on the rise among smaller companies. According to a recent Techaisle study, 18 percent of midmarket businesses are currently investing in big data and an additional 25 percent are planning to invest in these types of projects. Big data’s popularity, it seems, is bigger than ever.

But in many respects there is still much to learn about the right approach to big data. Conventional wisdom says that when tackling these initiatives, companies should first get their data in order so as to create a solid, integrated data foundation, and then acquire the right tools and software that can handle the analysis required.

But traveling the road from acquiring and organizing this influx of information to actually extracting meaningful insight through analytics requires more than just the right software. Big data is complex. Not only does it refer to large volumes of information, it can also come from a wide variety of sometimes uncertain sources, and often arrives at lightning speed. One of the most overlooked ingredients to embarking on these projects is having the right hardware platform that can handle the increased velocity and volume that can be too much for traditional systems to manage.

Big data needs more than commodity servers

Until now, those talking about big data have been focused solely on the latest analytics software that makes these activities possible. Ignoring the importance of the hardware piece of the puzzle, however, is bound to leave companies struggling.

Businesses that rely on commodity-type servers to handle their big data needs will find that they can be ill suited to meet the task. According to Anirban Chatterjee, Power Systems Product Marketing at IBM, “Commodity servers have a certain amount of data capability, but because they’re limited in how much of that capability they have, you often have to assemble a lot of servers together to get the amount of throughput you need and to work around server outages.”

Luckily, recent technological advances in hardware have made it easier for midsize companies to support the wide fluctuation of data flow and to subsequently analyze that data, sometimes within mere seconds. Chatterjee says, “Choosing the right platform is incredibly important because you can get a lot of benefit in how you handle data and eliminate restrictions on the access to your data so you can really transform it, modify it and analyze it much more effectively.”


"Platforms like Power Systems are designed specifically for the midmarket... Power is a much more efficient platform for big data, because it allows you to use a smaller solution that requires a lower amount of money up front."


Not unlike the hardware used within larger enterprises, the right hardware platform for smaller firms increasingly leverages virtualization and cloud-ready IT infrastructure. When dealing with high volumes of data, the platform chosen needs to manage diverse virtual resources while maximizing the productivity of the limited skills and resources that are inherent in midsize companies.

The challenge of limited resources has arguably pulled the plug on the big data efforts of many companies. The intensive computing demands of big data have strained commodity servers and, until recently, many IT budgets. However, the hardware options that are available today have evolved to better meet performance demands within a solution that is both easy to deploy and reasonably priced.

IBM Power Systems,TM for example, is built on an architecture that is designed around crunching data and moving it around very quickly. For midsize firms that are cost conscious, platforms like Power Systems are designed specifically for the midmarket. Chatterjee says, “Power is a much more efficient platform for big data, because it allows you to use a smaller solution that requires a lower amount of money up front. But you still get all of the hardware capability and platform capability that you need in order to tackle the data problem that you’re trying to solve.”

Chatterjee adds, “With a midsize company, the biggest constraint is their resources. They don’t have a whole team of people they can dedicate to bringing a data analysis solution online, doing the reports, doing the data analysis, developing the software to actually take their business data and pull insights out of it. So having a small turnkey solution gives them something they can just kind of plug in, turn on and have it automatically ingest the data and start spitting out the information that they need to know very, very quickly.”

Architecting for long-term success

As everything becomes more digitized, the amount of real-time data available to businesses will only grow. And as the volume and velocity of data coming into the organization are increased, scalability will be critical. Banding together commodity servers to meet big data needs may work in the short term, but companies need to anticipate the infrastructure that is needed for their long-term success when their big data efforts inevitably grow in scope and size. As in everything else in business, preparing for the future will be critical.

Subscrible Now

Related Content

in   f . Subscribe Now

Related Articles

  • Big data is quite simply data that cannot be managed or analyzed by traditional technologies.
    Read the article

  • You don't have to go back and reset the bar every time there's a new change.
    Read the article

  • Companies often think that their IT requirements are only going to expand as the business does.
    Read the article

  • Storage strategies that deliver business value.
    Read the article

  • Greener, leaner IT lowers costs, boosts efficiency.
    Read the article

Join the conversation

IBM for midsize business: news, events and more

Spotlight  

 

SMBs real-world use of big data