Data has been getting a lot of attention in the business world for a while now. First there was big data, which was another way to store data so that later the data could be analyzed. Recently the talk has been all about analyzing the data with new tools such as R and Python. The reality is that people who have been working with databases doing work in business intelligence have been analyzing data for a while. Learning a different toolset for analyzing data is not such a big leap, but an expansion of what they know.

As the field is rapidly expanding now, and demand is huge, now is a great time to learn the tools. With the advent of more advanced software, it seems essential to create fast and reliable processing mechanisms. In addition, concepts such as Web3 seem to be developing and becoming realities at an accelerated rate. Data plays an essential role in the development of such marvels of information technology. It is high time companies consider combining traditional data science tools, like MATLAB, Apache Spark, and SQL, with modern tools, like web3 sql. This could help them organize and process the data at relative speeds while reducing data analytics costs.

Traditional Data Science Development

Data scientist have created analysis solutions with data for a number of years. The data is analyzed, cleaned, processed with various algorithms, and results are created. When the process is complete, code has been created to provide meaning from a portion of the data and is ready to be migrated to production. Traditionally there has been a big gap between creating a solution and implementing the solution to be run against data on a regular basis. Data Scientists traditionally are not part of the IT organization, they are actuaries or analysts, not the people who have anything to do with system processing. Recently I did some work for a company and after the data scientists were done creating a solution, they turned over all of their code to the Java team. Six weeks later the code was released into production. This solution made no one happy. Management thought it took too long. The data scientist didn’t believe that the code that they created was what was implemented into production, and the java developers were tired of people blaming them for wrong code which required a long time to implement.

SQL Server Implementation of Data Science

Since SQL Server 2016 incorporates R and SQL Server 2017 has added the ability to include Python code into SQL Server, data science solutions can be incorporated as part of a scheduled process with SQL Server. There is now a dev ops solution for incorporating R and Python into SQL Server. One way of learning about the technology is through blogs and other online training which can help you get up to speed. Many times though there is no substitute for hands on learning. If you are attending PASS Summit 2017, and want to learn not only about data science, but how to incorporate it into SQL Server, I hope you can sign up for my all day training session on Applied Data Science for the SQL Server Professional. I hope to see you there.

I have recently created a You Tube channel where I plan on sharing more data related content where I have included my first video about this conference.

If you are at PASS Summit, please introduce yourself as I would love to meet people who read my blog personally.

 

Yours Always

Ginger Grant

Data aficionado et SQL Raconteur