TY - JOUR
T1 - ICE Buckets
T2 - Improved Counter Estimation for Network Measurement
AU - Einziger, Gil
AU - Fellman, Benny
AU - Friedman, Roy
AU - Kassner, Yaron
N1 - Funding Information:
Manuscript received May 4, 2016; revised January 27, 2018; accepted March 10, 2018; approved by IEEE/ACM TRANSACTIONS ON NETWORKING Editor S. Uhlig. Date of publication April 25, 2018; date of current version June 14, 2018. This work was supported in part by the Israeli Ministry of Science and Technology under Grant 3-10886 and in part by the Technion HPI Center. (Corresponding author: Yaron Kassner.) G. Einziger is with the Department of Electrical Engineering, Politecnico di Torino, 10129 Turin, Italy (e-mail: gilga1983@gmail.com).
Publisher Copyright:
© 1993-2012 IEEE.
PY - 2018/6/1
Y1 - 2018/6/1
N2 - Measurement capabilities are essential for a variety of network applications, such as load balancing, routing, fairness, and intrusion detection. These capabilities require large counter arrays in order to monitor the traffic of all network flows. While commodity SRAM memories are capable of operating at line speed, they are too small to accommodate large counter arrays. Previous works suggested estimators, which trade precision for reduced space. However, in order to accurately estimate the largest counter, these methods compromise the accuracy of the smaller counters. In this paper, we present a closed form representation of the optimal estimation function. We then introduce independent counter estimation buckets, a novel algorithm that improves estimation accuracy for all counters. This is achieved by separating the flows to buckets and configuring the optimal estimation function according to each bucket's counter scale. We prove a tighter upper bound on the relative error and demonstrate an accuracy improvement of up to 57 times on real Internet packet traces.
AB - Measurement capabilities are essential for a variety of network applications, such as load balancing, routing, fairness, and intrusion detection. These capabilities require large counter arrays in order to monitor the traffic of all network flows. While commodity SRAM memories are capable of operating at line speed, they are too small to accommodate large counter arrays. Previous works suggested estimators, which trade precision for reduced space. However, in order to accurately estimate the largest counter, these methods compromise the accuracy of the smaller counters. In this paper, we present a closed form representation of the optimal estimation function. We then introduce independent counter estimation buckets, a novel algorithm that improves estimation accuracy for all counters. This is achieved by separating the flows to buckets and configuring the optimal estimation function according to each bucket's counter scale. We prove a tighter upper bound on the relative error and demonstrate an accuracy improvement of up to 57 times on real Internet packet traces.
KW - Communications technology
KW - communication systems
KW - computer network management
KW - computer networks
KW - network security
UR - http://www.scopus.com/inward/record.url?scp=85045994536&partnerID=8YFLogxK
U2 - 10.1109/TNET.2018.2822734
DO - 10.1109/TNET.2018.2822734
M3 - Article
AN - SCOPUS:85045994536
SN - 1063-6692
VL - 26
SP - 1165
EP - 1178
JO - IEEE/ACM Transactions on Networking
JF - IEEE/ACM Transactions on Networking
IS - 3
ER -