While Brian's answer and Alexandros's answer have accurately described how to anonymize a submission and how to find out whether to do it, I would like to add an answer to the concrete question asked:
If so - what for, and why should I do this?
The assumed benefit of a double-blind review, in which the reviewers do not know who authored a given submission, is an improved fairness toward submitters.
Reviewers are supposed to judge a submission entirely and exclusively based on its contents. They should not take into account the reputation of the authors or their institution (or lack thereof), as verifiable and clearly described results are to be considered equally valuable coming from a reputable as well as coming from an unknown institution.
Likewise, while one can hope for reviewers to behave in a fair way, keeping submitters anonymous is another safety measure to prevent any personal biases based on culture, gender, or other personal factors from playing a role in the judgment on the submission.
Lastly, secondary effects of the judgment might be avoided. If submitters are known to reviewers, reviewers might have external incentives to provide a good or bad review. While direct conflicts of interest are generally excluded by review organizers (someone reviewing their own submission, for example), indirect interests of conflict (someone being promised a position in exchange for a favourable review; someone writing a bad review to prevent a direct competitor from publishing etc.) may not be apparent to those organizing reviews.
With all of this in mind, not all conferences conduct double-blind review for various reasons that are discussed in another Academia SE question.