Paper 2: To SQL, or not to SQL? Due: End of week 14 Now that you have been working on the Mapcushion project all semester and you have an understanding of NoSQL and the differences between NoSQL and relational databases. You have to apply this knowledge to a design problem. To date this semester you have been working with Mapcushion as a relational database, but that does not mean it might not also work as a NoSQL database. You must choose the data model you think is the best of the software and explain why. (NOTE: This does not in any way affect your project 2. You do not need to change how you are implementing the code for the project that continues on the way it is) 1. If you were the CIO or lead architect on the Mapcushion project, which data model (Relational, key-value, document, column-family or graph) would you choose and why? 2. What specific advantages of the model motivated your decision? 3. What specific disadvantages of other models motivated your decision? 4. Most importantly: Knowing that no system is perfect, what are the disadvantages associated with your CHOSEN model that you will need to mitigate? Your paper should be 750-1000 words. You may use citations as appropriate, but this is more of a technical paper describing your architecture decision. There is no absolute right or wrong choice in a data model! I am interested more in your explanation of why you made the choice than the actual model itself.