forked fromintel/hyperscan
- Notifications
You must be signed in to change notification settings - Fork5
Benchmarks
Ivan Koveshnikov edited this pageOct 27, 2022 ·6 revisions
CPU: 2x Intel(R) Xeon(R) Gold 6348 CPU @ 2.60GHz
NIC: 2x Intel(R) E810-2cqda2 (100G 2 ports)
PCIe: 4th gen
Total network bandwidth: 400 Gbps
Distro: Ubuntu 20.04 LTS
Kernel: 5.17
Driver: ice-1.9.11
CPU: 2x Intel(R) Xeon(R) Gold 6242R CPU @ 3.10GHz
NIC: 2x Intel(R) E810 (100G 1 of 2 ports connected)
PCIe: 3rd gen
Total network bandwidth: 200 Gbps
Distro: Ubuntu 20.04 LTS
Kernel: 5.17
Generator: Cisco T-Rex
We use 2 servers for the generation.
- Line rate - maximum practical throughput
- Base - base performance of XDP action, no regular expressions attached
- Literal - lookup for a constant sub-string inside of payload. Regex:
/private/s; Corpus: printable characters - Full payload scan - search with access to the whole payload. Regex
/pri.*ate/sH; Corpus: printable characters - Real-life - Our real-life regular expressions. 10 regexes in parallel, backtracking, search from the payload beginning.
We have evaluatedXDP_DROP andXDP_TX actions as primerally used in our network.
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 32.0 | 32.0 | 32.0 | 32.0 | 32.0 |
| 1000 | 48.0 | 48.0 | 48.0 | 48.0 | 48.0 |
| 512 | 88.0 | 88.0 | 88.0 | 88.0 | 88.0 |
| 256 | 176.0 | 162.0 | 165.0 | 108.0 | 161.0 |
| 128 | 328.0 | 307.0 | 244.0 | 108.0 | 208.0 |
| 64 | 568.0 | 430.0 | 257.0 | 211.0 | 220.0 |
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 384.0 | 384.0 | 384.0 | 384.0 | 384.0 |
| 1000 | 379.0 | 379.0 | 379.0 | 379.0 | 379.0 |
| 512 | 368.0 | 349.5 | 349.5 | 349.5 | 349.5 |
| 256 | 356.0 | 340.0 | 338.0 | 220.0 | 330.0 |
| 128 | 346.0 | 314.0 | 250.0 | 110.0 | 213.0 |
| 64 | 304.0 | 221.0 | 128.0 | 108.0 | 112.0 |
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 0% | 6% | 14% | 33% | 15% |
| 1000 | 0% | 8% | 20% | 48% | 22% |
| 512 | 0% | 16% | 34% | 80% | 38% |
| 256 | 0% | 28% | 68% | 98% | 74% |
| 128 | 0% | 51% | 98% | 98% | 98% |
| 64 | 0% | 75% | 98% | 98% | 99% |
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 32.0 | 32.0 | 32.0 | 32.0 | 32.0 |
| 1000 | 48.0 | 48.0 | 48.0 | 48.0 | 48.0 |
| 512 | 88.0 | 88.0 | 88.0 | 80.0 | 88.0 |
| 256 | 176.0 | 151.0 | 146.0 | 95.0 | 146.0 |
| 128 | 328.0 | 220.0 | 195.0 | 98.0 | 172.0 |
| 64 | 568.0 | 286.0 | 199.0 | 176.0 | 181.0 |
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 384.0 | 384.0 | 384.0 | 384.0 | 384.0 |
| 1000 | 379.0 | 379.0 | 379.0 | 379.0 | 379.0 |
| 512 | 368.0 | 349.5 | 349.5 | 323.0 | 349.5 |
| 256 | 356.0 | 310.0 | 299.0 | 196.0 | 300.0 |
| 128 | 346.0 | 226.0 | 200.0 | 100.0 | 176.0 |
| 64 | 304.0 | 146.0 | 102.0 | 89.0 | 101.0 |
| pkt size | linerate | base | literal | full payload scan | real-life |
|---|---|---|---|---|---|
| 1500 | 0% | 10% | 19% | 40% | 18% |
| 1000 | 0% | 12% | 28% | 56% | 26% |
| 512 | 0% | 22% | 48% | 98% | 45% |
| 256 | 0% | 36% | 75% | 98% | 86% |
| 128 | 0% | 53% | 98% | 98% | 99% |
| 64 | 0% | 83% | 98% | 98% | 99% |
BPF prog: 75%
Hyperscan: 28%
FPU load/store: 2.1%
BPF prog: 78%
Hyperscan: 40%
FPU load/store: 1.7%