I think it is essential to know Data Structures to be a successful Data Scientists. When looking at data, knowing that it’s Panel or Hierarchical Data immediately points you to what type of modeling techniques would work. For example, if you had a dataset of Patients, who all have Primary Care Physicians, Parents with children, Countries with different races, Companies with departments, and employees in each department, web traffic data with return visitors having session data etc., you will need ensure that you use Models that support these Hierarchial Data Structures. You can use regression, but you will need to do Multilevel regression that takes all levels into account. It’s also useful to know what kind of Data Structure you are dealing with if you need to partition variance at different levels.
The bottom line is, without a good understanding of your data, and knowing the structure is just one aspect of knowing, you can’t really have a full view to ensure you run appropriate modeling methods against that data set.
I think you see that it's not only required, but also useful to a Data Scientists to know Data Structures.