Some of the most common [SQL] queries I hear about from Big Data application developers are “How do I retrieve top n of x?“, “Find me last n of x?“, “How many times user u did blah?” This is especially true for large scale applications built around ad campaigns, user preferences and personalization. For example, in online advertising (AdTech) and online retail, companies make every effort to serve content that they think would be most interesting to you, and financially beneficial to them, based on a ton of data gathered from your previous visits. Let’s pause for a sec… when talking about querying Big Data, I am most certainly presuming that the data is denormalized and distributed. Right? Ok good, because I for one cannot imagine Big Data queries running against normalized data stored in traditional Relational Database Management Systems. Moving on… In this blog post, I will explore one such query and walk you through one way of data modeling and coding techniques using Aerospike’s open source NoSQL database—keeping in mind both storage and performance—to achieve the desired result. Prerequisites Node.js – To install latest stable version, click here Aerospike DB – To install the latest version, click here […]