Skip to main content

Table 2 Comparison between previous work that employed accelerators in the IP lookup problem. We present the algorithm used, the processor, and the maximum performance for IPv4 and IPv6. Finally, we also shown if the performance collected considered the PCIe data transfers, which are not necessary on our algorithm when using the KNL Intel Phi

From: High-performance IP lookup using Intel Xeon Phi: a Bloom filters based approach

Algorithm Computing device IPv4 (Mlps) IPv6 (Mlps) Considers PCIe transfer?
Radix Tree [7] GTX280 0.035 Yes
SAIL_L [8] Tesla C2075 547 Yes
Multi-bit Trie [9] Tesla C2075 2,900 3,600 No
DIR-24-8-BASIC (IPv4) 2x GTX480 76.17 74.22 Yes
Binary search (IPv6) [1]     Yes
GAMT [10] Tesla C2075 1,072 658 Yes
Bloomfwd Phi 7250 169.6 182.7