The COVID-19 pandemic has greatly affected all aspects of human life
including working of offices, businesses, industries, educational
institutions etc. With more work load shifting online, changes in the
network traffic are inevitable. The earlier investigations have
generally focused on the qualitative aspects of network traffic data
during COVID-19. In contrast, the paper presents a study based on MAWI
data characterizing network traffic in terms of multimodal and unimodal
probability distributions. It is found that a transition of multimodal
Gaussian mixture model of byte and packet counts during normal period to
that of unimodal Laplace distribution during COVID-19 period has
emerged. Further it is observed that the probability distribution
depicts the preponderance of small and large packets during normal
period which changes to that of small sized packets during Covid-19
period. These findings are likely to be useful to the administrators to
manage network during crisis periods.