SW-084 RFC 2544 Test Report
This page presents the RFC 2544 benchmark test results for the Brainboxes SW-084 5-port Gigabit Ethernet switch. RFC 2544 defines a standardised methodology for benchmarking network interconnect devices, covering throughput, latency, frame loss rate, and back-to-back frame handling.
Testing was performed using the Xena2544 v2.97 automated test suite on 28 January 2026, with a total test duration of 1 hour 33 minutes. Tests used custom frame sizes including standard and jumbo frames (64, 128, 256, 512, 1024, 1280, 1518, 2048, 4096, 9000, and 9018 bytes) in a full-mesh, bidirectional topology across all 5 ports at 1 Gbps.
The SW-084 passed all RFC 2544 tests. Throughput reached 99.6% of line rate for standard frames (64–4096 bytes) with zero frame loss. Jumbo frames at 9000 and 9018 bytes achieved reduced throughput (50% and 20%) as expected for the full-mesh 5-port topology.
| Test | Result |
|---|---|
| Throughput | Pass — 99.6% line rate for 64–4096 bytes; 50%/20% for 9000/9018 bytes |
| Latency & Jitter | Pass — at 50% and 100% rates |
| Frame Loss Rate | Pass — 0% at 50% rate (64–4096); minor loss at jumbo sizes |
| Back-to-Back | Pass — at 50% and 100% rates |
Test Configuration
| Parameter | Value |
|---|---|
| Topology | Mesh |
| Direction | Bidirectional |
| Frame Size Type | Custom Sizes |
| Frame Sizes | 64, 128, 256, 512, 1024, 1280, 1518, 2048, 4096, 9000, 9018 bytes |
| Port Count | 5 |
| Port Speed | 1 Gbps |
| Test Tool | Xena2544 v2.97 |
| Test Chassis | Abyssinian (S/N 2343912) |
Port Configuration
| Port ID | Interface Type | Speed | Rate |
|---|---|---|---|
| P-0-2-0 | SFP-E 10/100/1000M | Auto | 1 Gbps |
| P-0-2-1 | SFP-E 10/100/1000M | Auto | 1 Gbps |
| P-0-2-2 | SFP-E 10/100/1000M | Auto | 1 Gbps |
| P-0-2-3 | SFP-E 10/100/1000M | Auto | 1 Gbps |
| P-0-2-4 | SFP-E | 1 Gbit/s | 1 Gbps |
Throughput Test
The throughput test determines the maximum rate at which none of the offered frames are dropped by the device. Each frame size was tested for 30 seconds over 1 iteration, with a binary search from 0.1–100% line rate and 0% acceptable loss.
The SW-084 achieved 99.61% throughput for all standard frame sizes (64–4096 bytes) with zero frame loss. For jumbo frames, throughput was limited to 50.05% at 9000 bytes and 20.002% at 9018 bytes, reflecting the switch's internal bandwidth constraints at maximum jumbo frame sizes in a full 5-port mesh.
| Frame Size (bytes) | Offered Rate | Tx Rate (L1) | Tx Rate (L2) | Tx (Frames) | Rx (Frames) | Loss | Result |
|---|---|---|---|---|---|---|---|
| 64 | 99.61% | 4.980 Gbps | 3.795 Gbps | 222,343,200 | 222,343,200 | 0 | Pass |
| 128 | 99.61% | 4.980 Gbps | 4.307 Gbps | 126,194,400 | 126,194,400 | 0 | Pass |
| 256 | 99.61% | 4.980 Gbps | 4.620 Gbps | 67,669,200 | 67,669,200 | 0 | Pass |
| 512 | 99.61% | 4.980 Gbps | 4.793 Gbps | 35,106,600 | 35,106,600 | 0 | Pass |
| 1024 | 99.61% | 4.980 Gbps | 4.885 Gbps | 17,889,600 | 17,889,600 | 0 | Pass |
| 1280 | 99.61% | 4.980 Gbps | 4.904 Gbps | 14,366,400 | 14,366,400 | 0 | Pass |
| 1518 | 99.61% | 4.980 Gbps | 4.916 Gbps | 12,143,400 | 12,143,400 | 0 | Pass |
| 2048 | 99.61% | 4.980 Gbps | 4.932 Gbps | 9,031,200 | 9,031,200 | 0 | Pass |
| 4096 | 99.61% | 4.980 Gbps | 4.956 Gbps | 4,537,200 | 4,537,200 | 0 | Pass |
| 9000 | 50.05% | 2.501 Gbps | 2.496 Gbps | 1,039,800 | 1,039,800 | 0 | Pass |
| 9018 | 20.002% | 999.24 Mbps | 997.03 Mbps | 414,600 | 414,600 | 0 | Pass |
All 20 bidirectional port-to-port flows showed identical throughput with zero loss at every frame size.
Per-Port Throughput Results
All 5 ports produced identical results. Data shown is for port P-0-2-0.
Port P-0-2-0
| Frame Size (bytes) | Tx (Frames) | Tx Rate (L1) | Tx Rate (L2) | Tx Rate (fps) | Rx (Frames) |
|---|---|---|---|---|---|
| 64 | 44,468,640 | 996.10 Mbps | 758.93 Mbps | 1,482,288 | 44,468,640 |
| 128 | 25,238,880 | 996.09 Mbps | 861.49 Mbps | 841,296 | 25,238,880 |
| 256 | 13,533,840 | 996.09 Mbps | 923.91 Mbps | 451,128 | 13,533,840 |
| 512 | 7,021,320 | 996.09 Mbps | 958.64 Mbps | 234,044 | 7,021,320 |
| 1024 | 3,577,920 | 996.09 Mbps | 977.01 Mbps | 119,264 | 3,577,920 |
| 1280 | 2,873,280 | 996.07 Mbps | 980.75 Mbps | 95,776 | 2,873,280 |
| 1518 | 2,428,680 | 996.08 Mbps | 983.13 Mbps | 80,956 | 2,428,680 |
| 2048 | 1,806,240 | 996.08 Mbps | 986.45 Mbps | 60,208 | 1,806,240 |
| 4096 | 907,440 | 996.01 Mbps | 991.17 Mbps | 30,248 | 907,440 |
| 9000 | 207,960 | 500.21 Mbps | 499.10 Mbps | 6,932 | 207,960 |
| 9018 | 82,920 | 199.85 Mbps | 199.41 Mbps | 2,764 | 82,920 |
Ports P-0-2-1 through P-0-2-4 — identical results
All remaining 4 ports produced the same throughput, frame counts, and zero loss as P-0-2-0 above.
Latency and Jitter Test
The latency and jitter test measures forwarding delay and its variation across frame sizes. Tests were run for 1 second over 1 iteration using last-to-last latency mode, relative to the throughput test results. Testing was performed at both 50% and 100% of line rate.
Results at 50% Rate
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) | Result |
|---|---|---|---|---|---|---|
| 64 | 1.604 | 1.036 | 3.228 | 0.073 | 1.587 | Pass |
| 128 | 2.075 | 1.550 | 4.099 | 0.027 | 2.140 | Pass |
| 256 | 3.543 | 2.576 | 6.858 | 0.038 | 3.800 | Pass |
| 512 | 5.645 | 4.629 | 12.398 | 0.056 | 7.384 | Pass |
| 1024 | 10.347 | 8.734 | 22.614 | 0.066 | 13.494 | Pass |
| 1280 | 14.237 | 10.787 | 37.319 | 0.170 | 26.508 | Pass |
| 1518 | 16.611 | 12.695 | 34.176 | 0.070 | 21.432 | Pass |
| 2048 | 22.397 | 16.945 | 46.652 | 0.122 | 29.659 | Pass |
| 4096 | 38.741 | 33.366 | 84.819 | 0.197 | 47.123 | Pass |
| 9000 | 78.939 | 72.712 | 140.790 | 0.094 | 67.521 | Pass |
| 9018 | 73.347 | 72.832 | 136.580 | 0.122 | 63.167 | Pass |
Per-Port Latency and Jitter at 50% Rate
Port P-0-2-0
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) |
|---|---|---|---|---|---|
| 64 | 1.482 | 1.036 | 2.408 | 0.042 | 0.865 |
| 128 | 2.079 | 1.550 | 4.099 | 0.028 | 2.140 |
| 256 | 3.196 | 2.576 | 5.174 | 0.037 | 2.188 |
| 512 | 5.830 | 4.629 | 12.398 | 0.070 | 7.384 |
| 1024 | 10.093 | 8.734 | 22.614 | 0.058 | 13.494 |
| 1280 | 12.881 | 10.787 | 30.295 | 0.120 | 10.584 |
| 1518 | 15.530 | 12.719 | 32.203 | 0.065 | 12.316 |
| 2048 | 20.170 | 16.945 | 43.164 | 0.138 | 15.346 |
| 4096 | 34.754 | 33.366 | 77.122 | 0.229 | 32.979 |
| 9000 | 73.199 | 72.712 | 139.945 | 0.119 | 66.824 |
| 9018 | 73.153 | 72.856 | 73.289 | 0.015 | 0.072 |
Port P-0-2-4
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) |
|---|---|---|---|---|---|
| 64 | 2.009 | 1.569 | 3.228 | 0.130 | 1.587 |
| 128 | 2.204 | 2.082 | 3.116 | 0.024 | 0.986 |
| 256 | 3.708 | 3.108 | 5.369 | 0.036 | 2.188 |
| 512 | 5.695 | 5.161 | 8.937 | 0.046 | 3.728 |
| 1024 | 10.771 | 9.266 | 22.015 | 0.069 | 6.903 |
| 1280 | 14.372 | 11.319 | 29.047 | 0.124 | 17.391 |
| 1518 | 17.983 | 13.251 | 33.096 | 0.083 | 14.120 |
| 2048 | 22.958 | 17.501 | 41.508 | 0.100 | 16.718 |
| 4096 | 40.001 | 33.898 | 57.520 | 0.165 | 23.477 |
| 9000 | 73.330 | 73.220 | 140.790 | 0.088 | 67.521 |
| 9018 | 73.574 | 73.388 | 136.580 | 0.289 | 63.167 |
Ports P-0-2-1, P-0-2-2, P-0-2-3
Latency values for the remaining ports fell within the same range, with average latencies varying by less than ±20% across all 5 ports.
Results at 100% Rate
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) | Result |
|---|---|---|---|---|---|---|
| 64 | 1.985 | 1.036 | 2.961 | 0.016 | 1.250 | Pass |
| 128 | 2.912 | 1.550 | 4.271 | 0.045 | 2.044 | Pass |
| 256 | 3.929 | 2.576 | 6.861 | 0.021 | 2.213 | Pass |
| 512 | 7.759 | 4.629 | 12.976 | 0.024 | 4.281 | Pass |
| 1024 | 13.543 | 8.734 | 21.988 | 0.015 | 8.371 | Pass |
| 1280 | 15.104 | 10.787 | 24.065 | 0.022 | 8.347 | Pass |
| 1518 | 18.769 | 12.695 | 33.286 | 0.023 | 12.340 | Pass |
| 2048 | 24.521 | 16.945 | 37.993 | 0.019 | 15.707 | Pass |
| 4096 | 50.860 | 33.366 | 88.091 | 0.028 | 32.930 | Pass |
| 9000 | 89.514 | 72.712 | 145.192 | 0.120 | 72.116 | Pass |
| 9018 | 85.836 | 72.856 | 138.071 | 0.374 | 64.635 | Pass |
Per-Port Latency and Jitter at 100% Rate
Port P-0-2-0
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) |
|---|---|---|---|---|---|
| 64 | 1.982 | 1.036 | 2.576 | 0.015 | 0.673 |
| 128 | 2.458 | 1.550 | 3.209 | 0.065 | 1.250 |
| 256 | 3.907 | 2.576 | 5.222 | 0.021 | 1.323 |
| 512 | 7.539 | 4.653 | 10.282 | 0.025 | 2.020 |
| 1024 | 14.009 | 8.734 | 18.404 | 0.015 | 8.347 |
| 1280 | 15.585 | 10.787 | 19.759 | 0.020 | 8.298 |
| 1518 | 17.294 | 12.719 | 22.750 | 0.022 | 9.742 |
| 2048 | 22.258 | 16.969 | 26.134 | 0.019 | 8.731 |
| 4096 | 44.746 | 33.366 | 54.462 | 0.022 | 20.085 |
| 9000 | 88.738 | 72.712 | 118.849 | 0.085 | 46.040 |
| 9018 | 84.479 | 72.856 | 121.976 | 0.219 | 49.071 |
Port P-0-2-4
| Frame Size (bytes) | Avg Latency (us) | Min Latency (us) | Max Latency (us) | Avg Jitter (us) | Max Jitter (us) |
|---|---|---|---|---|---|
| 64 | 2.225 | 1.569 | 2.892 | 0.016 | 0.697 |
| 128 | 3.334 | 2.082 | 4.271 | 0.066 | 2.044 |
| 256 | 4.642 | 3.132 | 6.861 | 0.020 | 2.213 |
| 512 | 7.158 | 5.161 | 8.817 | 0.026 | 2.790 |
| 1024 | 14.594 | 9.290 | 18.624 | 0.016 | 8.371 |
| 1280 | 15.201 | 11.343 | 19.955 | 0.022 | 8.347 |
| 1518 | 15.869 | 13.227 | 19.145 | 0.024 | 5.725 |
| 2048 | 25.382 | 17.501 | 33.594 | 0.019 | 15.707 |
| 4096 | 51.202 | 33.922 | 64.616 | 0.038 | 29.467 |
| 9000 | 78.567 | 73.244 | 145.192 | 0.068 | 71.635 |
| 9018 | 88.914 | 73.388 | 138.071 | 0.312 | 64.635 |
Ports P-0-2-1, P-0-2-2, P-0-2-3
Latency values for the remaining ports fell within the same range as P-0-2-0 and P-0-2-4 above.
Frame Loss Rate Test
The frame loss rate test measures the percentage of frames that are not forwarded by the device under steady-state conditions. Each frame size was tested for 1 second over 1 iteration. No pass/fail criteria were configured.
At 50% rate, the SW-084 achieved zero frame loss for frame sizes 64–4096 bytes. Minor loss was observed at 9000 bytes (12 frames, 0.03%) and 9018 bytes (39 frames, 0.11%). At 100% rate, zero loss was seen for 64–1024 bytes, with increasing loss at larger jumbo frame sizes — reaching 29.52% at 9000 bytes and 31.19% at 9018 bytes at full line rate.
Results at 50% Rate
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Loss (Frames) | Loss Rate | Result |
|---|---|---|---|---|---|---|
| 64 | 3,720,220 | 3,720,220 | 2,500 Mbps | 0 | 0% | Pass |
| 128 | 2,111,480 | 2,111,480 | 2,500 Mbps | 0 | 0% | Pass |
| 256 | 1,132,240 | 1,132,240 | 2,500 Mbps | 0 | 0% | Pass |
| 512 | 587,400 | 587,400 | 2,500 Mbps | 0 | 0% | Pass |
| 1024 | 299,320 | 299,320 | 2,500 Mbps | 0 | 0% | Pass |
| 1280 | 240,380 | 240,380 | 2,500 Mbps | 0 | 0% | Pass |
| 1518 | 203,180 | 203,180 | 2,500 Mbps | 0 | 0% | Pass |
| 2048 | 151,100 | 151,100 | 2,500 Mbps | 0 | 0% | Pass |
| 4096 | 75,920 | 75,920 | 2,500 Mbps | 0 | 0% | Pass |
| 9000 | 34,640 | 34,628 | 2,500 Mbps | 12 | 0.03% | Pass |
| 9018 | 34,560 | 34,521 | 2,499 Mbps | 39 | 0.11% | Pass |
Per-Port Frame Loss at 50% Rate
All 5 ports showed zero loss for 64–4096 bytes. Minor loss at 9000 and 9018 bytes was distributed across specific port combinations.
Port P-0-2-0
| Frame Size (bytes) | Tx (Frames) | Tx Rate (L1) | Tx Rate (fps) | Rx (Frames) |
|---|---|---|---|---|
| 64 | 744,044 | 500 Mbps | 744,044 | 744,044 |
| 128 | 422,296 | 500 Mbps | 422,296 | 422,296 |
| 256 | 226,448 | 500 Mbps | 226,448 | 226,448 |
| 512 | 117,480 | 499.99 Mbps | 117,480 | 117,480 |
| 1024 | 59,864 | 499.98 Mbps | 59,864 | 59,864 |
| 1280 | 48,076 | 499.99 Mbps | 48,076 | 48,076 |
| 1518 | 40,636 | 499.99 Mbps | 40,636 | 40,636 |
| 2048 | 30,220 | 499.96 Mbps | 30,220 | 30,220 |
| 4096 | 15,184 | 499.98 Mbps | 15,184 | 15,184 |
| 9000 | 6,928 | 499.92 Mbps | 6,928 | 6,922 |
| 9018 | 6,912 | 499.77 Mbps | 6,912 | 6,912 |
Ports P-0-2-1 through P-0-2-4
All remaining ports showed zero loss for standard frames. At 9000 and 9018 bytes, individual port loss varied slightly (0–6 frames per port).
Results at 100% Rate
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Loss (Frames) | Loss Rate | Result |
|---|---|---|---|---|---|---|
| 64 | 7,440,460 | 7,440,360 | 5,000 Mbps | 100 | < 0.01% | Pass |
| 128 | 4,222,960 | 4,222,960 | 5,000 Mbps | 0 | 0% | Pass |
| 256 | 2,264,480 | 2,264,480 | 5,000 Mbps | 0 | 0% | Pass |
| 512 | 1,174,800 | 1,174,800 | 5,000 Mbps | 0 | 0% | Pass |
| 1024 | 598,640 | 598,640 | 5,000 Mbps | 0 | 0% | Pass |
| 1280 | 480,760 | 480,294 | 5,000 Mbps | 466 | 0.10% | Pass |
| 1518 | 406,360 | 406,066 | 5,000 Mbps | 294 | 0.07% | Pass |
| 2048 | 302,220 | 302,059 | 5,000 Mbps | 161 | 0.05% | Pass |
| 4096 | 151,840 | 151,834 | 5,000 Mbps | 6 | < 0.01% | Pass |
| 9000 | 69,280 | 48,828 | 4,999 Mbps | 20,452 | 29.52% | Pass |
| 9018 | 69,140 | 47,575 | 4,999 Mbps | 21,565 | 31.19% | Pass |
The high frame loss at 9000 and 9018 bytes at 100% rate is expected. The frame loss test sends at full line rate regardless of throughput results. Since the throughput test found the switch can only forward 9000-byte frames at ~50% rate across all 5 ports, sending at 100% exceeds the switch's forwarding capacity for jumbo frames. Standard frame sizes (64–4096 bytes) showed negligible loss.
Per-Port Frame Loss at 100% Rate
For standard frames, loss was zero or negligible across all ports. The significant jumbo frame loss at 100% rate was distributed across all ports, concentrated on port P-0-2-4 which received the highest loss.
Port P-0-2-0
| Frame Size (bytes) | Tx (Frames) | Tx Rate (L1) | Tx Rate (fps) | Rx (Frames) |
|---|---|---|---|---|
| 64 | 1,488,092 | 1,000 Mbps | 1,488,092 | 1,488,074 |
| 128 | 844,592 | 1,000 Mbps | 844,592 | 844,592 |
| 256 | 452,896 | 999.99 Mbps | 452,896 | 452,896 |
| 512 | 234,960 | 999.99 Mbps | 234,960 | 234,960 |
| 1024 | 119,728 | 999.97 Mbps | 119,728 | 119,728 |
| 1280 | 96,152 | 999.98 Mbps | 96,152 | 96,152 |
| 1518 | 81,272 | 999.97 Mbps | 81,272 | 81,272 |
| 2048 | 60,444 | 999.99 Mbps | 60,444 | 60,444 |
| 4096 | 30,368 | 999.96 Mbps | 30,368 | 30,367 |
| 9000 | 13,856 | 999.85 Mbps | 13,856 | 10,451 |
| 9018 | 13,828 | 999.82 Mbps | 13,828 | 9,283 |
Ports P-0-2-1 through P-0-2-4
All ports showed the same pattern: zero/negligible loss for standard frames, significant loss for 9000 and 9018 bytes at full rate.
Back-to-Back Test
The back-to-back test measures the maximum number of frames that can be sent at a given rate without any frame loss. Each frame size was tested with a 1-second burst over 1 iteration.
Results at 50% Rate
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) | Max Offered Rate (fps) | Max Offered Rate (Gbps) | Result |
|---|---|---|---|---|---|
| 64 | 3,720,220 | 238,094,080 | 3,720,220 | 2.50 | Pass |
| 128 | 2,111,480 | 270,269,440 | 2,111,480 | 2.50 | Pass |
| 256 | 1,132,240 | 289,853,440 | 1,132,240 | 2.50 | Pass |
| 512 | 587,400 | 300,748,800 | 587,400 | 2.50 | Pass |
| 1024 | 299,320 | 306,503,680 | 299,320 | 2.50 | Pass |
| 1280 | 240,380 | 307,686,400 | 240,380 | 2.50 | Pass |
| 1518 | 203,180 | 308,427,240 | 203,180 | 2.50 | Pass |
| 2048 | 151,100 | 309,452,800 | 151,100 | 2.50 | Pass |
| 4096 | 75,920 | 310,968,320 | 75,920 | 2.50 | Pass |
| 9000 | 9,740 | 87,660,000 | 34,640 | 1.25 | Pass |
| 9018 | 260 | 2,344,680 | 34,560 | 2.50 | Pass |
Per-Port Back-to-Back at 50% Rate
Burst frames were distributed equally across all 5 ports.
Port P-0-2-0
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) |
|---|---|---|
| 64 | 744,044 | 47,618,816 |
| 128 | 422,296 | 54,053,888 |
| 256 | 226,448 | 57,970,688 |
| 512 | 117,480 | 60,149,760 |
| 1024 | 59,864 | 61,300,736 |
| 1280 | 48,076 | 61,537,280 |
| 1518 | 40,636 | 61,685,448 |
| 2048 | 30,220 | 61,890,560 |
| 4096 | 15,184 | 62,193,664 |
| 9000 | 1,948 | 17,532,000 |
| 9018 | 52 | 468,936 |
Ports P-0-2-1 through P-0-2-4 — identical results
All remaining 4 ports produced the same burst frame counts as P-0-2-0 above.
Results at 100% Rate
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) | Max Offered Rate (fps) | Max Offered Rate (Gbps) | Result |
|---|---|---|---|---|---|
| 64 | 5,572,180 | 356,619,520 | 7,440,460 | 5.00 | Pass |
| 128 | 4,222,960 | 540,538,880 | 4,222,960 | 5.00 | Pass |
| 256 | 2,264,480 | 579,706,880 | 2,264,480 | 5.00 | Pass |
| 512 | 1,174,800 | 601,497,600 | 1,174,800 | 5.00 | Pass |
| 1024 | 598,640 | 613,007,360 | 598,640 | 5.00 | Pass |
| 1280 | 480,760 | 615,372,800 | 480,760 | 5.00 | Pass |
| 1518 | 284,520 | 431,901,360 | 406,360 | 5.00 | Pass |
| 2048 | 124,840 | 255,672,320 | 302,220 | 5.00 | Pass |
| 4096 | 76,200 | 312,115,200 | 151,840 | 5.00 | Pass |
| 9000 | 800 | 7,200,000 | 69,280 | 5.00 | Pass |
| 9018 | 800 | 7,214,400 | 69,140 | 5.00 | Pass |
Per-Port Back-to-Back at 100% Rate
Burst frames were distributed equally across all 5 ports. At 100% rate, jumbo frame burst sizes were very small (160 frames per port for 9000/9018 bytes).
Port P-0-2-0
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) |
|---|---|---|
| 64 | 1,114,436 | 71,323,904 |
| 128 | 844,592 | 108,107,776 |
| 256 | 452,896 | 115,941,376 |
| 512 | 234,960 | 120,299,520 |
| 1024 | 119,728 | 122,601,472 |
| 1280 | 96,152 | 123,074,560 |
| 1518 | 56,904 | 86,380,272 |
| 2048 | 24,968 | 51,134,464 |
| 4096 | 15,240 | 62,423,040 |
| 9000 | 160 | 1,440,000 |
| 9018 | 160 | 1,442,880 |
Ports P-0-2-1 through P-0-2-4 — identical results
All remaining 4 ports produced the same burst frame counts as P-0-2-0 above.
About This Test
RFC 2544 defines a standardised benchmarking methodology for network interconnect devices. The tests cover:
- Throughput — maximum rate at which no frames are lost
- Latency — forwarding delay at various frame sizes and rates
- Frame Loss Rate — percentage of frames lost under load
- Back-to-Back — maximum burst length the device can handle without loss
Testing was performed using the Xena2544 automated test suite with 1 iteration per test. This report extends the standard IEEE frame sizes with jumbo frames (2048, 4096, 9000, and 9018 bytes) to validate jumbo frame support. The reduced throughput at 9000 and 9018 bytes reflects the switch's forwarding capacity when handling maximum-size jumbo frames across all 5 ports simultaneously.