To categorize the data dimension of the problem, we look at its volume, velocity, and variety (the 3Vs), which are defined as follows:

• Volume: The volume is the expected size of the data that the algorithm will process.
• Velocity: The velocity is the expected rate of new data generation when the algorithm is used. It can be zero.
• Variety: The variety quantifies how many different types of data the designed algorithm is expected to deal with.

The following figure shows the 3Vs of the data in more detail. The center of this diagram shows the simplest possible data, with a small volume and low variety and velocity. As we move away from the center, the complexity of the data increases. It can increase in one or more of the three dimensions. For example, in the dimension of velocity, we have the Batch process as the simplest, followed by the Periodic process, and then the Near Real-Time process. Finally, we have the Real-Time process, which is the most complex to handle in the context of data velocity. For example, a collection of live video feeds gathered by a group of monitoring cameras will have a high volume, high velocity, and high variety and may need an appropriate design to have the ability to store and process data effectively. On the other hand, a simple .csv file created in Excel will have a low volume, low velocity, and low variety:

For example, if the input data is a simple csv file, then the volume, velocity, and variety of the data will be low. On the other hand, if the input data is the live stream of a security video camera, then the volume, velocity, and variety of the data will be quite high and this problem should be kept in mind while designing an algorithm for it.