The emergence of cheaper sensors that can be connected in sensor networks and to the Internet has led to the concept of Internet of Things (IoT). By 2020 the estimate is that there will be tens of billions of data-spouting devices (sensors) that will be connected to the internet. This new technology trend will result in dramatic increase to Big data with high volumes and velocities. These data-spouting devices continuously capture, store, analyze and send data to cloud. These tiny devices are found everywhere today for example smoke detectors, smart cars, door locks, industrial robots, street lights, heart monitors, trains, wind turbines even in tennis racquets and toasters. Traditional databases are typically not created to deal with the Big data aspects. Although database vendors tried to handle and store large volume of data with better performance but in the last decade, dramatically increase in size of data caused several issues to handle and store the data with faster access and less storage cost. NoSQL databases are considered to be an important component of Big Data when it comes to retrieving and storing large amount of data with reduced storage cast. Companies and organizations tend to find a need to scale up quickly and efficiently to cover the dramatically increase in the data volume and demands of services they create. Most important factor to be considered is to find a cost-efficient solution that can meet the demands and requirements. CITI-SENSE (project for air and environmental quality, 2012-2016) sensor data and ProaSense (project of oil and gas drilling) sensor data are considered as reference point to deal with and to store, evaluate and benchmark the performance using MongoDB (NoSQL), RDF (Graph Database), WFS and traditional SQL database solutions. This report investigates the different database solutions available today. What do SQL, NoSQL and Graph databases offer? Comparison and differentiation? What are their pros and cons? The purpose of this database benchmark and evaluation study is to figure out which database solution shows better performance for CRUD operations (Create, Read (Query), Update, Delete) with respect to response time, cost, storage capacity and scalability for sensor data.