In my younger days (we won’t discuss how long ago that was), I decided to take up sailing. I had always been fascinated watching sailboats, gliding smoothly along the water, with nothing but wind as their engine. My first lesson on the water was an exhilarating and eye opening experience. The instructor took me out on a 17 foot day sailer. It was a brisk New England April afternoon, what seemed like a perfect day to be sailing. The instructor explained the different techniques and maneuvers required to traverse the lake, which I executed, though somewhat nervously. We were moving at what felt like breakneck speeds, constantly shifting and changing the position of the boat, adjusting the sail to the changing wind directions, while still heading in the desired direction. Suddenly, a gust of wind came from a completely unexpected direction. Before I fully realized what had happened, I found myself treading water, looking back at the day sailer, which was completely turtled (a term I learned that day) upside down in the water. My instructor, amazingly, was standing on top of the upside down boat, completely dry!
Today in technology, we are trying to sail great lakes of data. It is estimated that we are creating 2.5 quintillion bytes (a billion gigabytes) of data daily. That number is going to continue to grow with the explosion of the Internet of Things. IT organizations are struggling with helping the business to sail and navigate these lakes, leveraging the power they present, providing value back to the business. What steps can the IT organization take to avoid ending up treading water in those lakes? What can be done to be like my sailing instructor, who remained dry and safe on top of the boat?
Big data without analytics is just taking up storage space
With all the hype around the concept of ‘big data’ and the sheer volume of information being created on a daily basis, many organizations are engaging in ‘data hoarding’. These organizations are capturing and storing every little piece of data available to them. The assumption is that the data will provide some intrinsic value just by having it. Data without an understanding of the meaning is just occupying space. This understanding is more than just knowing the business, though that knowledge is required. Properly analyzing these huge volumes requires deep technical skills in computer science, communication, and statistics. Without leveraging these skillsets, business runs the risk of introducing false correlations from the analysis. (For some entertaining examples of false correlations check out www.tylervigen.com).
False correlations can have negative business impact; decisions being made based on incorrect assumptions can lead to potentially costly errors. To avoid treading in those dangerous waters, the IT organization should ensure they are providing the appropriate resources for both the storage and analysis of these great data lakes. Working closely with the business, providing those resources, explaining the risks and benefits involved will be a big step in staying dry. It’s not the only step nor the only area risk.
Security, privacy, archiving – not all data is created equal
In the world of data, there has always been the challenge to balance the needs of the business, user convenience, while also addressing privacy, security and regulatory concerns. Everything is a tradeoff; there is no perfect way of striking that balance. We are now in a new world, where we are creating great big data lakes of information, and we are developing new uses for this data at breakneck speeds. Frequently, this is done before there is a solid understanding of the implications and tradeoffs involved. A breach or compliance violation can be like that gust of wind that turtled my sailboat, catching the organizations unaware, leaving them, like me, treading water in the great big data lake.
The IT organization has a responsibility to help the business understand the tradeoffs and risks involved in this rapidly changing environment. Not all data is created equal. Personal information is very different from statistics obtained from the census bureau. There is no one-size-fits-all for dealing with privacy issues, security issues, or compliance. Some of the data is sourced internally, some sourced from external systems, including possibly social media systems. What responsibility does the organization have for privacy around externally sourced data? If it passes through your system, resides in your system, regardless of original source, you have a responsibility to handle the data appropriately. Policies and procedures need to be defined and executed.
Depending on the industry, there are various regulatory and compliance rules around retention and archiving of data. Even with the constant influx of new data, the IT organization is responsible for ensuring compliance. Another consideration often overlooked, is leveraging archiving rules around data retention help deal with privacy and security concerns. Once data has been analyzed and rolled up, consider if the raw data is still required. Archiving or rolling raw data off systems when no longer required can help deal with privacy and security concerns around transitory data sourced from other systems. Data can’t be compromised if it’s no longer in the system. This can be a difficult sell to the business, as there is a tendency to hoard data, its human nature. It’s the classic ‘I might need it again’. As already mentioned, everything is a tradeoff. A well-defined archiving strategy can help strike the appropriate balance.
Smooth sailing across the data lakes requires good planning and design
After we returned to shore from my first disastrous attempt at sailing, I asked the instructor how he managed to be on top of the boat completely dry. He explained that all the instructors go through training and planning for just such scenarios. They would actually force the boats to turtle, and practice climbing over the boat in the opposite direction as it flipped. He also told me he knew the flip would be coming. Looking across the lake, he saw ripples in the water that signaled a coming change in wind direction, allowing him time to react quickly to the anticipated shift.
No technology negates the need for good planning and design. It’s not just about storing structured and unstructured data. It’s not just providing the latest and greatest analytic tools. For the IT organization to successfully sail the great big data lakes, they too must plan and design how to leverage and balance the data and its analysis. They must provide the appropriate resources working with the business to ensure there is the correct understanding of the data that is available. Those that succeed will be like my instructor, standing triumphantly on top of the boat. Those that do not will be like I was, soaking wet, treading water, and wondering what had happened.