Presenter: YingYing Chen
Studies of search Engines such as Google and Bing have reported on tremendous revenue losses with the increase in the delay. For instance 0.5 sec of delay in Google searches result in the drop of 20% revenue. This paper studies the variations in the search response time (SRT) in peak and off-peak hours. The authors found counter intuitive results that off-peak hours result in high delay.
To find out the different factors that lead to higher SRT in off-peak hours, this paper further investigate major impact factors such as servers, network, browsers, and query. They perform ANOVA to decompose the variations in the different time intervals. Their results show that more than 65% of the variance was due to network factor, followed by variations due to the browser speed and the nature of the query.
Server side processing time has relatively small contribution to the overall delay.
Studies of search Engines such as Google and Bing have reported on tremendous revenue losses with the increase in the delay. For instance 0.5 sec of delay in Google searches result in the drop of 20% revenue. This paper studies the variations in the search response time (SRT) in peak and off-peak hours. The authors found counter intuitive results that off-peak hours result in high delay.
To find out the different factors that lead to higher SRT in off-peak hours, this paper further investigate major impact factors such as servers, network, browsers, and query. They perform ANOVA to decompose the variations in the different time intervals. Their results show that more than 65% of the variance was due to network factor, followed by variations due to the browser speed and the nature of the query.
Server side processing time has relatively small contribution to the overall delay.
Higher network latencies are due to more queries from the residential networks in the off-peak hours. In addition, authors found that off-peak hours have different nature of queries in terms of average number of images requested as compare to the queries sent from the enterprise network. Overall, it was argued that understanding SRT is challenging due to changes in the user demographics that lead to systematic variations in the SRT. Performance debugging in SRT is tained by the user behavior.
Q1: Do you focus on only browser requests or have you also considered mobile
apps.
Ans: We have only considered browsers requests.
q2: (Dina P.) : how do you compute your ground truth for your three techniques
Ans: We compared it against the tickets generator by the operators.
Q2: (Paul Barford) There are many papers on time-series based techniques to identify the anomalies. Did you consider standard time-series or wavelet based methods.
No.
(Comment from Paul) When you show your technique then by default you argue your technique against other techniques.
This comment has been removed by the author.
ReplyDeleteTheir results show that more than 65% of the variance was due to network factor, followed by variations due to the browser speed and the nature of the query. iPage coupons
ReplyDelete