SlamData CTO John De Goes on Forbes: Have Your Cake And Eat It: Big Data Without Hadoop.
From the article:
Today’s NoSQL databases share many characteristics with Hadoop, but in some cases, they are easier to manage and develop for. So if you’re about to embark on a big data project, it makes sense to investigate at lease the leading NoSQL contenders. Because for data hubs, IoT and real-time analytics uses cases, one of these may be a wiser choice than a Hadoop-built solution.
“But there are a few use cases where NoSQL databases are stronger. Among them are:
Data Hubs: If you need to consolidate information from a variety of sources of data (particularly online data), NoSQL databases are a perfect fit. Common examples of this include creating a single view of customer or patient data, or linking together data from lots of different departments inside a larger organization. The insurance company MetLife uses a NoSQL database in this fashion, and I have worked with several others companies doing the same.
IoT: The Internet of Things is rife with devices and sensors that create large amounts of data whose structure can change and whose primary purpose is rollup for monitoring, alerting and large-scale analytics. NoSQL databases excel at this and can handle incredible volumes of data with ease. Engineering firm Bosch is one company using NoSQL database technology for IoT.
Real-Time Analytics: Most NoSQL databases have the ability to perform real-time aggregation on streams of data (for example, clickstream or log data), which can make them an excellent choice for high-volume, low-value data that needs to be pre-aggregated before analysis. Content delivery networks Buffer and MaxCDN both leverage NoSQL tech for real-time analytics.”
Read the entire article on Forbes: Have Your Cake And Eat It: Big Data Without Hadoop
John is part of the Forbes Technology Council, an invitation-only organization comprised of elite CIOs, CTOs and technology executives.