Posted By: David Malmborg
FairCom hits the road with its 2016 Data Strategies Roadshow to demonstrate how enterprises can handle the changing environment of data.
Throughout the early part of 2016, FairCom will be on the road with its Data Roadshow. Last week we held our first meeting in Austin, Texas with Bloor Group. It was a great turnout with some illuminating presentations given by Gilbert Van Cutsem, Evaldo Oliveira, and Dr. Robin Bloor of the Bloor Group.
Catching the Next Wave or Staying Constant in the Data-driven Economy
Gilbert Van Cutsem is Consultant and Advisor, NoSQL & Multimodel Database Technology, and previously the Senior Vice President and General Manager, Database Division at Pervasive Software. He gave a presentation titled “Catching the Next Wave or Constant in the Age of the Data-Driven Economy.”
Looking back at the golden age of our industry, the common thread that runs through recent history is data, and by extension, databases. There are many parallels to be drawn between the gold rush and the ongoing data rush. Companies and enterprises today are built around the data they collect and produce. And it is this data that will define the future of the company.
While most of the time you hear the business analyst talking about the economy, the truth is that such analysis handles the data as post-facto information that sustains them. The current data revolution is forcing this to change. Data has become larger, faster, and more diverse, enriching all economic analysis in ways nobody would expected a few years ago. In Gilbert’s opinion, we’re not in the era of economy anymore, but rather in the era where economy is directly influenced and lead by data. Rather than being a simple passive component of the analysis, the rapid changes imposed by Big Data transformed this relationship, and data is actually driving the worldwide economy today.
Calling this the data-driven economy, enterprises today are working to capture and digitize all the data and build their data footprint. Data is converted into value when it can be used in real-time to make business decisions and increase revenue and profits. More and more organizations are working extensively to find better ways to monetize their data—new and legacy. They are being challenged by the complexity of Big Data’s 4V: velocity, variety, volume, and veracity. Technology vendors need to keep this in mind when bringing solutions to the market.
There are three driving factors that are forcing organizations to rethink how they handle the coming deluge of data. Robin Bloor, Ph.D., laid out these three factors in his presentation as the evolving technology landscape, the growing “data story,” and the evolution and revolution of database technology.
When looking at the technology landscape today, you will quickly realize that “we are going through the most vast changes in IT that have ever occurred.” Hardware has increased quicker and faster with CPUs becoming processing clusters; memory becoming the primary store for data; and SSDs replacing disks while living up to the same Moore’s law as CPUs. These changes alone are dramatic. But software has also changed and its disruption is keeping up with hardware. The open source software and business model is working. Distributed architecture is becoming more prevalent and allowing for more real-time oriented software. This is allowing companies to become more data-driven—better said, they are becoming more real-time data-driven.
The next factor has to be the growing “data story” as Dr. Bloor calls it. Corporate data volumes grow at about 55% annually. This exponential growth has been going on at this rate for maybe 40 years now. When it comes to big data, Dr. Bloor said, “it’s not really about big data as much as it about big analytics.” Meaning that even though companies are producing a large volume of data every year, their ability to analyze that data is becoming more important today. However, according to Bloor, there aren’t enough data scientists out there for companies, and many businesses are looking at ways to automate this data analysis.
The final factor to look at when setting your data strategy is to understand the evolution—or revolution—of database technology. The deluge of data means many different data structures exist today. Traditional databases were never built to handle data structure flexibility; hence many types of database are required. The term “best-fit engineering” is catching on to help companies identify database technology that is the “best-fit” for their engineering needs and to not rely solely on the “general” big names in the DBMS market.
Harmonizing Your Unstructured Data With SQL Capabilities
With the growing shift in data demands within enterprise organizations today, Evaldo took a minute to discuss how to handle these demands with the confluence of NoSQL and SQL.
It’s hard to predict the future (although Big Analytics is changing this), and the world of data has been no different. One thing that remains consistent with data predictions is the need for NoSQL. The NoSQL market seems to be growing at 40-60% annually, with an expected $4.2B market in 2020. “NoSQL DBMSs are no longer fringe technologies,” according to Nick Heudecker, Merv Adrian (OPDBMS analysts). “They have become substantial differentiators for many enterprises.” NoSQL has been a strong option for enterprises with massive velocity and a wide variety of data. However, much of the enterprise is entrenched in SQL-oriented data. So do you have to choose SQL vs. NoSQL?
The question being asked today is “how can you harmonize your existing SQL data with your NoSQL needs?” One of the growing options is the use of a multimodel DBMS. A multimodel database is designed to support multiple data models against a single, integrated backend. Document, graph, relational, and key-value models are examples of data models that may be supported by a multimodel database.
The new c-treeACE V11 offers a variety of ways to enable enterprise organizations to have SQL with their unstructured data. Designed for high-throughput transactions, c-treeACE gives you ACID transactions with NoSQL data as well as full SQL access to that same data for analysis purposes. In the presentation embedded below, you can see a few different ways in which c-treeACE is able to handle different data models with the same data.
In the end, predicting the future is hard, and sometimes is not recommended. Instead, look at your current dataset and try to adjust it to the new needs with focus on your business goals. Figuring out how to combine SQL and NoSQL is more feasible than it looks, and a multimodel DBMS is the technology to support you.