With Great Speed Come Small Buffers: Space-Bandwidth Tradeoffs for Routing

被引:2
作者
Miller, Avery [1 ]
Patt-Shamir, Boaz [2 ]
Rosenbaum, Will [3 ]
机构
[1] Univ Manitoba, Dept Comp Sci, Winnipeg, MB, Canada
[2] Tel Aviv Univ, Sch Elect Engn, Tel Aviv, Israel
[3] Max Planck Inst Informat, Saarbrucken, Germany
来源
PROCEEDINGS OF THE 2019 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING (PODC '19) | 2019年
基金
加拿大自然科学与工程研究理事会; 以色列科学基金会;
关键词
packet networks; adversarial queuing theory; packet forwarding; buffer overflow; multiple destinations; directed paths; directed trees; DELAY;
D O I
10.1145/3293611.3331614
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We consider the Adversarial Queuing Theory (AQT) model, where packet arrivals are subject to a maximum average rate 0 <= rho <= 1 and burstiness sigma >= 0. In this model, we analyze the size of buffers required to avoid overflows in the basic case of a path. Our main results characterize the space required by the average rate and the number of distinct destinations: we show that O(ld(1/l) + sigma) space suffice, where d is the number of distinct destinations and l = left perpendicular1/rho left perpendicular; and we show that Omega( 1/ld(1/l) + sigma) space is necessary. For directed trees, we describe an algorithm whose buffer space requirement is at most 1+ d' + sigma where d' is the maximum number of destinations on any root-leaf path.
引用
收藏
页码:117 / 126
页数:10
相关论文
共 20 条
[1]  
Adler M., 2002, STACS 2002. 19th Annual Symposium on Theoretical Aspects of Computer Science. Proceedings (Lecture Notes in Computer Science Vol.2285), P88
[2]  
Aiello W, 2003, SIAM PROC S, P771
[3]   General dynamic routing with per-packet delay guarantees of O(distance+1/session rate) [J].
Andrews, M ;
Fernández, A ;
Harchol-Balter, M ;
Leighton, T ;
Zhang, L .
SIAM JOURNAL ON COMPUTING, 2000, 30 (05) :1594-1623
[4]  
Awerbuch B., 1993, Proceedings. 34th Annual Symposium on Foundations of Computer Science (Cat. No.93CH3368-8), P32, DOI 10.1109/SFCS.1993.366884
[5]   Instability of FIFO at arbitrarily low rates in the adversarial queueing model [J].
Bhattacharjee, R ;
Goel, A ;
Lotker, Z .
SIAM JOURNAL ON COMPUTING, 2004, 34 (02) :318-332
[6]   Adversarial queuing theory [J].
Borodin, A ;
Kleinberg, J ;
Raghavan, P ;
Sudan, M ;
Williamson, DP .
JOURNAL OF THE ACM, 2001, 48 (01) :13-38
[7]   The Design of Competitive Online Algorithms via a Primal Dual Approach [J].
Buchbinder, Niv ;
Naor, Joseph .
FOUNDATIONS AND TRENDS IN THEORETICAL COMPUTER SCIENCE, 2007, 3 (2-3) :93-263
[8]   A CALCULUS FOR NETWORK DELAY .1. NETWORK ELEMENTS IN ISOLATION [J].
CRUZ, RL .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1991, 37 (01) :114-131
[9]   Optimal Local Buffer Management for Information Gathering with Adversarial Traffic [J].
Dobrev, Stefan ;
Lafond, Manuel ;
Narayanan, Lata ;
Opatrny, Jaroslav .
PROCEEDINGS OF THE 29TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES (SPAA'17), 2017, :265-274
[10]  
Even Guy, 2016, 24 ANN EUR S ALG ESA, V57, DOI 10.4230/LIPIcs.ESA.2016.40