New Year! New Thoughts on NoSQL
2014 has kicked off with a bang. The Polar Vortex gave weather groupies a chance to use big data to see how low their temps could go and why. Google snapped up (if a $3.2B acquisition can be viewed as “snapped up") one of the leaders in connected home/“the Internet of things.” And everyone has their predictions for the new year. While I don’t necessarily have any announcements similar to the Polar Vortex or the Nest acquisition, I do have my predictions for NoSQL for 2014.
Looking Back is Looking Forward
NoSQL access layers in 2013 began to standardize. Many software vendors streamlined their access layers/APIs to lower the barriers to entry for new technologists and implementations. I think this trend will continue as we move into 2014 and will start to take on the look of the early SQL language development days.
If you set the “waay back machine” to the late 80s and early 90s, you would see differing implementations of the SQL language (aka API or access layer) by different vendors and different groups. I believe that the process of evolving SQL into the lingua franca of data access will serve as a model for NoSQL. Different NoSQL platforms will continue to develop their own access layers, but we will start to see a similar process to SQL’s development for NoSQL platforms, starting with
NoSQL Will Start to Venture Into the World of SQL
A few of the major SQL database vendors and some of the smaller ones have begun to integrate NoSQL features into their SQL languages. Providing access to JSON data sets/formats appears to be the initial step toward this integration and the target of many standardization efforts. JSON is both popular and a relatively safe formatting standard to use. So it provides low hanging fruit for the SQL data store vendors to target. Yet, this shows the power of the “SQL lobby” to flex to meet data management challenges/challengers to the throne for data access.
Integrating JSON “dot notation” into standard SQL interfaces shows that multi-structured data sets are not just for NoSQL platforms. The XML-lite format has many strong supporters in multiple vertical industries as a fast and effective way to store more information that was easily accessible in traditional formats. Storing multi-structured datasets and integrating them into existing SQL-based RDBMS processing workloads will be an important driving factor in 2014. And it will be the start of putting NoSQL access in SQL.
NoSQL Will Make Major Strides in the World of Business Intelligence and Analytics
Somewhere along the line, SQL data access and structure data formats became synonymous with business intelligence and the practice of analytics. This probably comes from the fact that SQL data stores and structured data is all that BI had before. However, that doesn’t mean it has to continue.
With the development of NoSQL access standards and the blurring of lines between SQL and NoSQL mentioned above, multi-structured data sets will start to be accepted in traditional BI implementations. Also, new BI (as well as data discovery) implementations will be built on top of mutli-structured data stores and access layers. The dominance of SQL/RDBMS in BI will start to erode in 2014 and possibly be in full fledge decline in 2015.
NOTE: This is not to say that SQL’s death certificate has been post-dated and/or your data warehouse will be magically transported to MongoDB or Hadoop Hive on January 1, 2015. No, but NoSQL will start to have a much stronger presence and acceptance in BI and that trend will continue.
What say the readers?
- Are you spitting Red Bull on the keyboard/tablet at the “sound” of NoSQL following SQL’s development path?
- Are you excited to see access to JSON data types in traditional RDBMs?
- Did you get the urge to go laugh at the BICC’s team knowing that they won’t like sharing with the NoSQL guys?
Provide your comments below and/or ping me via twitter at @JohnLMyers44 with the hashtag #noodlingNoSQL.