Title
Measurement based analysis, modeling, and synthesis of the internet delay space
Abstract
Understanding the characteristics of the Internet delay sp ace (i.e., the all-pairs set of static round-trip propagation delays a mong edge networks in the Internet) is important for the design of global-scale distributed systems. For instance, algorithms used in over lay net- works are often sensitive to violations of the triangle ineq uality and to the growth properties within the Internet delay space. Si nce de- signers of distributed systems often rely on simulation and emula- tion to study design alternatives, they need a realistic mod el of the Internet delay space. Our analysis shows that existing models do not adequately cap- ture important properties of the Internet delay space. In th is paper, we analyze measured delays among thousands of Internet edge net- works and identify key properties that are important for distributed system design. Furthermore, we derive a simple model of the In- ternet delay space based on our analytical findings. This mod el preserves the relevant metrics far better than existing models, al- lows for a compact representation, and can be used to synthesize delay data for simulations and emulations at a scale where direct measurement and storage are impractical.
Year
DOI
Venue
2006
10.1145/1177080.1177091
IEEE/ACM Transactions on Networking
Keywords
Field
DocType
system design,analysis,simulation,internet edge network,static round-trip propagation delay,delay data,measurement,realistic model,design alternative,important property,simple model,internet delay space,modeling,distributed system,edge network,synthesis,performance
Network delay,Distributed system design,Computer science,Internet traffic engineering,Computer network,Emulation,Triangle inequality,Overlay network,The Internet
Conference
Volume
Issue
ISBN
18
1
1-59593-561-4
Citations 
PageRank 
References 
74
3.17
27
Authors
6
Name
Order
Citations
PageRank
Bo Zhang11336.74
T. S. Eugene Ng22491274.31
Animesh Nandi3114076.25
Rudolf H. Riedi448539.83
Peter Druschel58715861.36
Guohui Wang6108860.78