Application and Comparison of Data Plane Acceleration Technologies for NFV 


Vol. 42,  No. 8, pp. 1636-1646, Aug.  2017
10.7840/kics.2017.42.8.1636


PDF
  Abstract

In this paper, we compared and analyzed the performance of common network technology and packet-acceleration techniques such as SR-IOV, DPDK, FD.io which is used for accelerating data planes in an OpentSack based NFVI environment. we use two compute node and one controller and the packet generator (Send & Recive) is configured to measure the network performance. we deployed two kinds of VNF(non-vEPC,vEPC) and meassure the performance between the VM(VM to VM). Experimental results showed that SR-IOV is the highest network performance in common network environment and DPDK performance is similar to SR-IOV in NF structure such as vEPC using Raw Ethernet. FD.io was the highest performance in Phy2Phy (Physical-vSwitch-Physical) state, but the case of VM2VM showed that the performance was degraded due to logical port bottleneck between VM and vSwitch.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

Y. Park, H. Yang, Y. Kim, "Application and Comparison of Data Plane Acceleration Technologies for NFV," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 8, pp. 1636-1646, 2017. DOI: 10.7840/kics.2017.42.8.1636.

[ACM Style]

Young-ki Park, Hyun-sik Yang, and Young-han Kim. 2017. Application and Comparison of Data Plane Acceleration Technologies for NFV. The Journal of Korean Institute of Communications and Information Sciences, 42, 8, (2017), 1636-1646. DOI: 10.7840/kics.2017.42.8.1636.

[KICS Style]

Young-ki Park, Hyun-sik Yang, Young-han Kim, "Application and Comparison of Data Plane Acceleration Technologies for NFV," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 8, pp. 1636-1646, 8. 2017. (https://doi.org/10.7840/kics.2017.42.8.1636)