TY - GEN
T1 - Implementation of WLRU algorithm to improve scalability in software defined network
AU - Nurwarsito, Heru
AU - Nareindra, Aryadna
N1 - Publisher Copyright:
© 2020 ACM.
PY - 2020/11/16
Y1 - 2020/11/16
N2 - Packet forwarding in a Software Defined Network (SDN) architecture was conducted by a matching process between packet information with flow entry. Network traffic with multiple IP or MAC addresses will increase the number of flow entry insertion and may result in the flow table on the switch running out of space for new flow entry. Flow table overflow that occurred will decrease network scalability in packet forwarding. This study implemented the Weighted Least Recently Used (WLRU) algorithm in the flow table's entry management process to prevent flow table overflow conditions and reduce the number of repeated insertions for an entry. The implementation was conducted using a single linked list data structure and Snort as a packet monitoring. The entry management implementations using the WLRU algorithm successfully overcame the flow table overflow conditions and improved network scalability by being able to forward up to 402 packets, which means 4 times more than before, that only 97 packets. In trade-off overcame the flow table overflow condition, the WLRU algorithm affected the RTT delay value with a minimum value of 19.54 milliseconds, a maximum value of 1607.40 milliseconds, and an average value 195.72 milliseconds.
AB - Packet forwarding in a Software Defined Network (SDN) architecture was conducted by a matching process between packet information with flow entry. Network traffic with multiple IP or MAC addresses will increase the number of flow entry insertion and may result in the flow table on the switch running out of space for new flow entry. Flow table overflow that occurred will decrease network scalability in packet forwarding. This study implemented the Weighted Least Recently Used (WLRU) algorithm in the flow table's entry management process to prevent flow table overflow conditions and reduce the number of repeated insertions for an entry. The implementation was conducted using a single linked list data structure and Snort as a packet monitoring. The entry management implementations using the WLRU algorithm successfully overcame the flow table overflow conditions and improved network scalability by being able to forward up to 402 packets, which means 4 times more than before, that only 97 packets. In trade-off overcame the flow table overflow condition, the WLRU algorithm affected the RTT delay value with a minimum value of 19.54 milliseconds, a maximum value of 1607.40 milliseconds, and an average value 195.72 milliseconds.
KW - flow entry
KW - flow table
KW - software defined network (SDN)
KW - WLRU
UR - https://www.scopus.com/pages/publications/85099060250
U2 - 10.1145/3427423.3427444
DO - 10.1145/3427423.3427444
M3 - Conference contribution
AN - SCOPUS:85099060250
T3 - ACM International Conference Proceeding Series
SP - 165
EP - 170
BT - Proceedings of 2020 International Conference on Sustainable Information Engineering and Technology, SIET 2020
PB - Association for Computing Machinery
T2 - 5th International Conference on Sustainable Information Engineering and Technology, SIET 2020
Y2 - 16 November 2020 through 17 November 2020
ER -