Different kinds of networks at different levels of system design have evolved in the last decade, mainly riding on top of global Internet. Regardless of the type of the network, these networks can be viewed as content sharing and distribution network. Understanding the popularity dynamics of the contents, termed here as Content Hotness, is useful in many ways including the characterization of the workload as well as the system design and evaluation. Despite the fact that popularity skewness among contents has been well studied, the temporal dynamics of popularity of a given content has not been studied extensively. We attempt to propose a discrete time Markov chain (DTMC) model to model such content level hotness dynamics. We focus in two realistic scenario and see how such model can be used to represent the temporal variation in the content popularity.
[1]
SzaboGabor,et al.
Predicting the popularity of online content
,
2010
.
[2]
Seungjoon Lee,et al.
Modeling channel popularity dynamics in a large IPTV system
,
2009,
SIGMETRICS '09.
[3]
Sandip Roy,et al.
The influence model
,
2001
.
[4]
Songqing Chen,et al.
The stretched exponential distribution of internet media access patterns
,
2008,
PODC '08.
[5]
Herwig Bruneel,et al.
Analysis and Modeling of Video Popularity Evolution in Various Online Video Content Systems: Power-Law versus Exponential Decay
,
2009,
2009 First International Conference on Evolving Internet.
[6]
Bernardo A. Huberman,et al.
Predicting the popularity of online content
,
2008,
Commun. ACM.