If I have 100k visitors to my site, I send test A to all 100k, and test B to all 100k does that take 100k of my traffic allocation or 200k?
You can find a good explanation on following article, here.
However, setting the traffic allocation for your experiment to 100% of your website visitors (100k) and setting each variation (A+B) to each 50%, leads to a distribution of 50k visitors to variation A and 50k visitors to varariation B.
In case your question is related to how unique vistors are counted while running two experiments at the same time, please have a look at this article.
I hope that answers your question.
EMEA | Amsterdam