PE-515 RFC 2544 Test Report
The Brainboxes PE-515 5-port embedded Gigabit Ethernet switch was benchmarked in-house against RFC 2544, the industry-standard methodology for network device performance testing. Testing was performed on 2026-06-01 over 14 hours 18 minutes using the Xena2544 v2.97 automated test suite across all 5 ports in a full-mesh, bidirectional topology at 1 Gbps per port (5 Gbps aggregate).
Unlike the standard RFC 2544 sweep, this run used a custom frame-size set: the seven IEEE default sizes (64–1518 bytes) extended with four jumbo sizes — 2048, 4096, 8192 and 9000 bytes — for eleven sizes in total.
The PE-515 passed all four RFC 2544 sub-tests — Throughput, Latency/Jitter, Frame Loss Rate, and Back-to-Back — at every one of the eleven frame sizes, including all four jumbo sizes up to 9000 bytes. Throughput reached 4.980 Gbps at Layer 1 with zero loss, essentially the full 5 Gbps aggregate line rate, and held there from minimum-length 64-byte frames all the way to 9000-byte jumbo frames.
| Test | Result |
|---|---|
| Throughput | Pass |
| Latency and Jitter | Pass |
| Frame Loss Rate | Pass |
| Back-to-Back | Pass |
Two figures on this page are easy to misread, and both are expected behaviour rather than defects:
- The throughput test converged to 99.61%, not 100%. That is an artefact of the test's 0.500% binary-search resolution, not of measured loss — loss was zero at every frame size at the converged rate.
- The frame-loss test forces a fixed 100% offered rate, above that converged ceiling. Under that small overload the switch dropped roughly 0.001% of frames at most sizes, rising to 0.0121% at 9000 bytes. At 50% rate loss was zero everywhere. See Frame Loss Rate Test.
Test Configuration
| Parameter | Value |
|---|---|
| Customer Service ID | PE-515rev4 |
| Test Date | 2026-06-01, 23:14 |
| Test Duration | 14:18:26 (h:m:s) |
| Test Tool | Xena2544 v2.97, running on Microsoft Windows 10 Enterprise |
| Chassis | C-0 — Abyssinian (serial 2343912), 192.168.0.39:22606 |
| Module | M-0-6 — Odin-10G-5S-6P-CU (serial 650338) |
| Topology | Mesh |
| Direction | Bidirectional |
| Frame Size Type | Custom Sizes (not IEEE Default) |
| Frame Sizes | 64, 128, 256, 512, 1024, 1280, 1518, 2048, 4096, 8192, 9000 bytes |
| Port Count | 5 |
| Port Speed | AUTO negotiation, 1 G port rate per port |
| Aggregate Capacity | 5 Gbps (5 ports × 1 Gbps) |
| Bidirectional Flows | 20 (5 ports × 4 peers) |
| Iterations per Test | 3 |
| Test Duration per Iteration | 60 seconds |
| Rates Tested (L/J, FL, B2B) | 50.00% and 100.00% of line rate |
Port Configuration
| Port ID | Interface | Speed | Rate |
|---|---|---|---|
| P-0-6-1 | 10GBASE-T | AUTO | 1 G |
| P-0-6-2 | 10GBASE-T | AUTO | 1 G |
| P-0-6-3 | 10GBASE-T | AUTO | 1 G |
| P-0-6-4 | 10GBASE-T | AUTO | 1 G |
| P-0-6-5 | 10GBASE-T | AUTO | 1 G |
10GBASE-T is the interface type of the Xena test module's ports, not of the PE-515. Each test port was rate-limited to 1 G with auto-negotiation to match the PE-515's Gigabit ports. The PE-515 is a 5-port 10/100/1000 Mbps switch. All five ports sit on one 6-port Odin module; module port 0 was unused, so port numbering starts at P-0-6-1.
Throughput Test
The throughput test determines the maximum offered rate at which the device under test can forward frames with zero loss. The test ran for 60 seconds per trial across 3 iterations, starting at an initial rate of 100.000% of line rate and binary-searching between a maximum of 100.000% and a minimum of 0.000% with a resolution of 0.500%. Acceptable loss was set to 0.0000% — zero frame loss permitted — and the rate result scope was Common Result, so a single throughput rate is reported across all five ports.
The PE-515 converged to 99.61% offered rate at every frame size, with zero loss throughout. Peak aggregate throughput was 4.980 Gbps at Layer 1 — effectively the full 5 Gbps line rate of five Gigabit ports — sustained from 64 bytes through 8192 bytes and easing only fractionally to 4.979 Gbps at 9000 bytes.
Layer 2 throughput climbs from 3.795 Gbps at 64 bytes to 4.968 Gbps at 9000 bytes. That rise is expected arithmetic, not a change in switch behaviour: each frame carries a fixed 20 bytes of preamble and inter-frame gap, so as frames grow that overhead shrinks as a proportion of the wire time. Layer 1 utilisation stays flat at ~4.980 Gbps throughout, which means the switch saturates its backplane at every frame size — including minimum-length 64-byte frames, the hardest case for any switch.
| 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 | 444,684,000 | 444,684,000 | 0 | Pass |
| 128 | 99.61% | 4.980 Gbps | 4.307 Gbps | 252,387,600 | 252,387,600 | 0 | Pass |
| 256 | 99.61% | 4.980 Gbps | 4.620 Gbps | 135,338,400 | 135,338,400 | 0 | Pass |
| 512 | 99.61% | 4.980 Gbps | 4.793 Gbps | 70,213,200 | 70,213,200 | 0 | Pass |
| 1024 | 99.61% | 4.980 Gbps | 4.885 Gbps | 35,779,200 | 35,779,200 | 0 | Pass |
| 1280 | 99.61% | 4.980 Gbps | 4.904 Gbps | 28,732,800 | 28,732,800 | 0 | Pass |
| 1518 | 99.61% | 4.980 Gbps | 4.916 Gbps | 24,286,800 | 24,286,800 | 0 | Pass |
| 2048 | 99.61% | 4.980 Gbps | 4.932 Gbps | 18,062,400 | 18,062,400 | 0 | Pass |
| 4096 | 99.61% | 4.980 Gbps | 4.956 Gbps | 9,074,400 | 9,074,400 | 0 | Pass |
| 8192 | 99.61% | 4.980 Gbps | 4.968 Gbps | 4,548,000 | 4,548,000 | 0 | Pass |
| 9000 | 99.61% | 4.979 Gbps | 4.968 Gbps | 4,140,000 | 4,140,000 | 0 | Pass |
All 20 bidirectional flows (5 ports × 4 destinations each) showed zero loss at every frame size, with per-flow received frames matching transmitted frames exactly.
Per-Port Throughput Results
All five ports are byte-identical at every frame size — same frame counts, same L1 and L2 rates, same frames per second — so no port is favoured or starved. Port P-0-6-1 is shown as representative.
Port P-0-6-1
| Frame Size (bytes) | Tx (Frames) | Tx Rate (L1) | Tx Rate (L2) | Tx Rate (fps) | Rx (Frames) |
|---|---|---|---|---|---|
| 64 | 88,936,800 | 996.09 Mbps | 758.93 Mbps | 1,482,280 | 88,936,800 |
| 128 | 50,477,520 | 996.09 Mbps | 861.48 Mbps | 841,292 | 50,477,520 |
| 256 | 27,067,680 | 996.09 Mbps | 923.91 Mbps | 451,128 | 27,067,680 |
| 512 | 14,042,640 | 996.09 Mbps | 958.64 Mbps | 234,044 | 14,042,640 |
| 1024 | 7,155,840 | 996.09 Mbps | 977.01 Mbps | 119,264 | 7,155,840 |
| 1280 | 5,746,560 | 996.07 Mbps | 980.75 Mbps | 95,776 | 5,746,560 |
| 1518 | 4,857,360 | 996.08 Mbps | 983.13 Mbps | 80,956 | 4,857,360 |
| 2048 | 3,612,480 | 996.08 Mbps | 986.45 Mbps | 60,208 | 3,612,480 |
| 4096 | 1,814,880 | 996.01 Mbps | 991.17 Mbps | 30,248 | 1,814,880 |
| 8192 | 909,600 | 995.95 Mbps | 993.53 Mbps | 15,160 | 909,600 |
| 9000 | 828,000 | 995.81 Mbps | 993.60 Mbps | 13,800 | 828,000 |
Ports P-0-6-2 through P-0-6-5 — identical results at every frame size.
Latency and Jitter Test
The latency and jitter test measures one-way frame transit time at a given offered rate. Each frame size was tested for 60 seconds across 3 iterations using Last-to-Last latency mode, relative to the throughput-determined rate. In the summary tables, average latency and average jitter are the mean across all 5 ports; minimum and maximum are the worst case observed on any port. The report also records a minimum jitter row, which is 0.000 µs on every port at every frame size, so it is omitted here.
Results at 50% Rate
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) | Result |
|---|---|---|---|---|---|---|
| 64 | 1.74 | 1.48 | 3.21 | 0.03 | 1.18 | Pass |
| 128 | 2.35 | 1.99 | 4.09 | 0.04 | 1.88 | Pass |
| 256 | 3.63 | 3.01 | 8.11 | 0.03 | 5.08 | Pass |
| 512 | 6.28 | 5.06 | 14.1 | 0.02 | 6.99 | Pass |
| 1024 | 10.9 | 9.17 | 27.6 | 0.03 | 13.3 | Pass |
| 1280 | 13.0 | 11.2 | 37.9 | 0.04 | 26.6 | Pass |
| 1518 | 16.5 | 13.2 | 40.4 | 0.04 | 26.3 | Pass |
| 2048 | 25.8 | 17.4 | 56.8 | 0.05 | 30.2 | Pass |
| 4096 | 39.4 | 33.7 | 80.5 | 0.05 | 46.7 | Pass |
| 8192 | 77.2 | 66.5 | 192.8 | 0.12 | 117.8 | Pass |
| 9000 | 80.4 | 73.0 | 192.7 | 0.10 | 95.2 | Pass |
At half line rate the switch forwards a 64-byte frame in 1.74 µs on average and a 9000-byte jumbo frame in 80.4 µs, with average jitter never exceeding 0.12 µs at any frame size. Worst-case maximum latency peaks at 192.8 µs on 8192-byte frames — marginally above the 192.7 µs recorded at 9000 bytes. That inversion is not an error: at large frame sizes the maximum is a rare tail event driven by queue coincidence rather than by frame length, so it does not rise strictly with frame size the way the average does.
Per-Port Latency and Jitter at 50% Rate
Port P-0-6-1
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.672 | 1.479 | 2.602 | 0.038 | 0.785 |
| 128 | 2.395 | 1.993 | 4.085 | 0.040 | 1.563 |
| 256 | 3.834 | 3.011 | 7.221 | 0.032 | 3.167 |
| 512 | 5.864 | 5.064 | 11.951 | 0.025 | 6.807 |
| 1024 | 10.832 | 9.169 | 27.635 | 0.038 | 11.482 |
| 1280 | 12.372 | 11.214 | 24.388 | 0.043 | 10.511 |
| 1518 | 17.515 | 13.178 | 40.384 | 0.049 | 14.745 |
| 2048 | 25.412 | 17.356 | 47.151 | 0.045 | 24.591 |
| 4096 | 37.854 | 33.729 | 71.567 | 0.052 | 35.023 |
| 8192 | 67.501 | 66.508 | 165.437 | 0.129 | 65.797 |
| 9000 | 80.640 | 72.970 | 145.110 | 0.091 | 72.076 |
Port P-0-6-2
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.708 | 1.504 | 2.530 | 0.034 | 0.857 |
| 128 | 2.425 | 2.017 | 4.037 | 0.039 | 1.563 |
| 256 | 3.552 | 3.035 | 6.916 | 0.034 | 2.277 |
| 512 | 6.042 | 5.088 | 11.959 | 0.022 | 6.719 |
| 1024 | 11.125 | 9.193 | 23.369 | 0.039 | 11.546 |
| 1280 | 12.593 | 11.238 | 21.477 | 0.044 | 10.199 |
| 1518 | 18.345 | 13.218 | 39.574 | 0.046 | 26.251 |
| 2048 | 23.968 | 17.380 | 48.066 | 0.053 | 17.936 |
| 4096 | 37.609 | 33.753 | 77.757 | 0.047 | 43.082 |
| 8192 | 75.667 | 66.524 | 184.360 | 0.138 | 117.788 |
| 9000 | 79.215 | 72.994 | 192.731 | 0.105 | 87.294 |
Port P-0-6-3
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.756 | 1.496 | 3.211 | 0.032 | 1.042 |
| 128 | 2.356 | 2.009 | 3.500 | 0.036 | 1.371 |
| 256 | 3.489 | 3.027 | 6.138 | 0.032 | 2.910 |
| 512 | 6.611 | 5.080 | 14.068 | 0.027 | 5.845 |
| 1024 | 11.632 | 9.185 | 20.010 | 0.032 | 10.383 |
| 1280 | 13.010 | 11.230 | 28.878 | 0.041 | 17.559 |
| 1518 | 16.638 | 13.210 | 38.917 | 0.048 | 15.635 |
| 2048 | 25.907 | 17.372 | 47.713 | 0.042 | 16.846 |
| 4096 | 37.576 | 33.745 | 80.515 | 0.041 | 46.690 |
| 8192 | 86.097 | 66.516 | 192.827 | 0.136 | 65.805 |
| 9000 | 80.905 | 72.994 | 167.409 | 0.100 | 72.212 |
Port P-0-6-4
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.818 | 1.479 | 2.738 | 0.033 | 1.106 |
| 128 | 2.331 | 1.993 | 3.933 | 0.036 | 1.884 |
| 256 | 3.445 | 3.019 | 6.555 | 0.031 | 3.279 |
| 512 | 6.762 | 5.064 | 14.036 | 0.024 | 6.991 |
| 1024 | 10.842 | 9.169 | 22.576 | 0.028 | 8.403 |
| 1280 | 14.046 | 11.222 | 37.930 | 0.041 | 26.644 |
| 1518 | 15.045 | 13.202 | 30.329 | 0.036 | 17.046 |
| 2048 | 27.687 | 17.364 | 56.821 | 0.056 | 30.164 |
| 4096 | 42.133 | 33.737 | 69.506 | 0.054 | 35.673 |
| 8192 | 81.407 | 66.524 | 167.698 | 0.114 | 65.741 |
| 9000 | 84.655 | 73.002 | 171.619 | 0.105 | 72.276 |
Port P-0-6-5
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.746 | 1.488 | 2.730 | 0.025 | 1.178 |
| 128 | 2.225 | 2.001 | 3.516 | 0.032 | 1.323 |
| 256 | 3.808 | 3.027 | 8.111 | 0.031 | 5.083 |
| 512 | 6.136 | 5.072 | 11.863 | 0.019 | 6.679 |
| 1024 | 9.900 | 9.177 | 22.656 | 0.025 | 13.334 |
| 1280 | 13.051 | 11.222 | 29.038 | 0.038 | 10.776 |
| 1518 | 14.953 | 13.202 | 30.706 | 0.040 | 17.479 |
| 2048 | 25.946 | 17.364 | 50.303 | 0.042 | 24.591 |
| 4096 | 41.756 | 33.745 | 75.119 | 0.047 | 41.358 |
| 8192 | 75.286 | 66.516 | 167.642 | 0.093 | 65.805 |
| 9000 | 76.471 | 72.986 | 168.307 | 0.090 | 95.240 |
Results at 100% Rate
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) | Result |
|---|---|---|---|---|---|---|
| 64 | 1.92 | 1.48 | 2.76 | 0.03 | 0.72 | Pass |
| 128 | 2.74 | 1.99 | 3.88 | 0.03 | 1.16 | Pass |
| 256 | 4.48 | 3.02 | 8.09 | 0.04 | 2.25 | Pass |
| 512 | 7.82 | 5.06 | 13.5 | 0.07 | 4.38 | Pass |
| 1024 | 14.7 | 9.17 | 30.6 | 0.04 | 8.48 | Pass |
| 1280 | 17.5 | 11.2 | 35.7 | 0.04 | 10.5 | Pass |
| 1518 | 21.4 | 13.2 | 39.8 | 0.02 | 12.4 | Pass |
| 2048 | 27.1 | 17.4 | 55.6 | 0.02 | 16.7 | Pass |
| 4096 | 49.5 | 33.7 | 99.1 | 0.02 | 33.0 | Pass |
| 8192 | 105.7 | 66.5 | 224.8 | 0.03 | 65.7 | Pass |
| 9000 | 119.5 | 73.0 | 261.7 | 0.04 | 72.3 | Pass |
At full rate, average latency rises against the 50% figures by up to about half again at large frame sizes — 119.5 µs versus 80.4 µs at 9000 bytes, a 1.49× increase — which is the expected cost of deeper queueing when every port is saturated. Small frames barely move: 1.92 µs versus 1.74 µs at 64 bytes. Average jitter never exceeds 0.07 µs at any frame size; it is lower than at 50% rate for 1518-byte frames and above, and marginally higher between 256 and 1024 bytes. Aggregate transmitted and received frame counts match exactly at every frame size in this test.
Note that this test's offered rate is 99.609% — it runs relative to the throughput result rather than at a literal 100% of line rate.
Per-Port Latency and Jitter at 100% Rate
Port P-0-6-1
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.854 | 1.479 | 2.426 | 0.022 | 0.697 |
| 128 | 2.689 | 1.993 | 3.268 | 0.035 | 1.018 |
| 256 | 4.807 | 3.019 | 8.054 | 0.041 | 2.253 |
| 512 | 7.928 | 5.072 | 13.499 | 0.020 | 4.305 |
| 1024 | 14.934 | 9.169 | 27.354 | 0.116 | 8.475 |
| 1280 | 19.814 | 11.222 | 35.733 | 0.041 | 10.447 |
| 1518 | 19.495 | 13.186 | 30.602 | 0.021 | 11.947 |
| 2048 | 25.648 | 17.364 | 39.013 | 0.016 | 16.597 |
| 4096 | 46.174 | 33.745 | 68.977 | 0.023 | 32.906 |
| 8192 | 121.748 | 66.508 | 224.788 | 0.028 | 65.709 |
| 9000 | 132.889 | 72.986 | 261.728 | 0.032 | 72.188 |
Port P-0-6-2
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.935 | 1.504 | 2.426 | 0.023 | 0.721 |
| 128 | 2.783 | 2.017 | 3.877 | 0.035 | 1.154 |
| 256 | 4.184 | 3.035 | 6.018 | 0.042 | 2.188 |
| 512 | 7.961 | 5.088 | 11.190 | 0.020 | 4.305 |
| 1024 | 13.474 | 9.193 | 22.247 | 0.011 | 7.553 |
| 1280 | 18.697 | 11.238 | 28.004 | 0.041 | 10.455 |
| 1518 | 21.915 | 13.194 | 37.506 | 0.022 | 12.372 |
| 2048 | 28.378 | 17.380 | 55.611 | 0.020 | 16.597 |
| 4096 | 48.234 | 33.761 | 79.641 | 0.022 | 32.971 |
| 8192 | 108.385 | 66.524 | 179.878 | 0.031 | 65.725 |
| 9000 | 124.515 | 72.994 | 191.737 | 0.032 | 72.172 |
Port P-0-6-3
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 2.033 | 1.496 | 2.762 | 0.045 | 0.721 |
| 128 | 2.746 | 2.009 | 3.628 | 0.036 | 1.138 |
| 256 | 4.378 | 3.027 | 6.066 | 0.035 | 2.245 |
| 512 | 7.558 | 5.080 | 9.883 | 0.021 | 4.281 |
| 1024 | 15.161 | 9.185 | 25.206 | 0.016 | 8.387 |
| 1280 | 15.883 | 11.230 | 24.692 | 0.038 | 10.367 |
| 1518 | 23.576 | 13.218 | 39.735 | 0.021 | 12.372 |
| 2048 | 27.990 | 17.372 | 53.630 | 0.017 | 16.581 |
| 4096 | 51.856 | 33.753 | 79.625 | 0.026 | 32.890 |
| 8192 | 102.408 | 66.516 | 148.622 | 0.029 | 65.733 |
| 9000 | 105.144 | 72.994 | 179.709 | 0.043 | 72.276 |
Port P-0-6-4
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.879 | 1.479 | 2.538 | 0.021 | 0.721 |
| 128 | 2.724 | 1.993 | 3.380 | 0.031 | 1.146 |
| 256 | 4.233 | 3.019 | 7.132 | 0.034 | 2.237 |
| 512 | 7.672 | 5.064 | 13.347 | 0.136 | 4.369 |
| 1024 | 14.931 | 9.185 | 29.383 | 0.047 | 8.467 |
| 1280 | 16.141 | 11.222 | 29.383 | 0.034 | 10.439 |
| 1518 | 22.038 | 13.202 | 39.767 | 0.023 | 12.372 |
| 2048 | 28.333 | 17.364 | 45.347 | 0.018 | 16.653 |
| 4096 | 50.557 | 33.753 | 99.118 | 0.023 | 33.035 |
| 8192 | 91.927 | 66.508 | 146.931 | 0.029 | 65.725 |
| 9000 | 113.164 | 72.978 | 175.363 | 0.034 | 72.188 |
Port P-0-6-5
| Frame Size (bytes) | Avg Latency (µs) | Min Latency (µs) | Max Latency (µs) | Avg Jitter (µs) | Max Jitter (µs) |
|---|---|---|---|---|---|
| 64 | 1.889 | 1.488 | 2.402 | 0.021 | 0.705 |
| 128 | 2.751 | 2.001 | 3.709 | 0.026 | 1.162 |
| 256 | 4.806 | 3.027 | 8.087 | 0.032 | 2.237 |
| 512 | 7.992 | 5.072 | 11.607 | 0.142 | 4.377 |
| 1024 | 15.219 | 9.177 | 30.594 | 0.025 | 8.467 |
| 1280 | 16.960 | 11.222 | 25.350 | 0.029 | 10.431 |
| 1518 | 19.798 | 13.202 | 36.712 | 0.019 | 12.332 |
| 2048 | 25.397 | 17.372 | 37.032 | 0.020 | 16.525 |
| 4096 | 50.437 | 33.745 | 76.113 | 0.023 | 32.906 |
| 8192 | 103.994 | 66.516 | 164.410 | 0.024 | 65.741 |
| 9000 | 121.891 | 72.994 | 213.699 | 0.036 | 72.228 |
Frame Loss Rate Test
The frame loss test counts dropped frames at fixed offered rates. Each frame size was tested for 60 seconds across 3 iterations at 50% and 100% of line rate. The test executed with Use Pass/Fail Criteria: No — all values are reported informationally and every frame size was marked PASS.
At 50% rate the PE-515 forwarded every frame at every size with zero loss. At a forced 100% rate it dropped a small but non-zero number of frames at every frame size — see below.
Results at 50% Rate
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Loss (Frames) | Loss Rate | Result |
|---|---|---|---|---|---|---|
| 64 | 223,213,200 | 223,213,200 | 2.500 Gbps | 0 | 0.00% | Pass |
| 128 | 126,688,800 | 126,688,800 | 2.500 Gbps | 0 | 0.00% | Pass |
| 256 | 67,934,400 | 67,934,400 | 2.500 Gbps | 0 | 0.00% | Pass |
| 512 | 35,244,000 | 35,244,000 | 2.500 Gbps | 0 | 0.00% | Pass |
| 1024 | 17,959,200 | 17,959,200 | 2.500 Gbps | 0 | 0.00% | Pass |
| 1280 | 14,422,800 | 14,422,800 | 2.500 Gbps | 0 | 0.00% | Pass |
| 1518 | 12,190,800 | 12,190,800 | 2.500 Gbps | 0 | 0.00% | Pass |
| 2048 | 9,066,000 | 9,066,000 | 2.500 Gbps | 0 | 0.00% | Pass |
| 4096 | 4,555,200 | 4,555,200 | 2.500 Gbps | 0 | 0.00% | Pass |
| 8192 | 2,282,400 | 2,282,400 | 2.499 Gbps | 0 | 0.00% | Pass |
| 9000 | 2,078,400 | 2,078,400 | 2.500 Gbps | 0 | 0.00% | Pass |
Per-Port Frame Loss at 50% Rate
All five ports are identical, each offering ~500 Mbps. Port P-0-6-1 is shown as representative.
Port P-0-6-1
| Frame Size (bytes) | Tx (Frames) | Tx Rate (L1) | Tx Rate (fps) | Rx (Frames) |
|---|---|---|---|---|
| 64 | 44,642,640 | 500.00 Mbps | 744,044 | 44,642,640 |
| 128 | 25,337,760 | 500.00 Mbps | 422,296 | 25,337,760 |
| 256 | 13,586,880 | 500.00 Mbps | 226,448 | 13,586,880 |
| 512 | 7,048,800 | 499.99 Mbps | 117,480 | 7,048,800 |
| 1024 | 3,591,840 | 499.98 Mbps | 59,864 | 3,591,840 |
| 1280 | 2,884,560 | 499.99 Mbps | 48,076 | 2,884,560 |
| 1518 | 2,438,160 | 499.99 Mbps | 40,636 | 2,438,160 |
| 2048 | 1,813,200 | 499.96 Mbps | 30,220 | 1,813,200 |
| 4096 | 911,040 | 499.98 Mbps | 15,184 | 911,040 |
| 8192 | 456,480 | 499.82 Mbps | 7,608 | 456,480 |
| 9000 | 415,680 | 499.92 Mbps | 6,928 | 415,680 |
Ports P-0-6-2 through P-0-6-5 — identical results, zero loss at every frame size.
Results at 100% Rate
The Loss Rate column shows the value the report prints, followed in parentheses by the true percentage computed as Loss ÷ Tx. The printed 0% is a rounding artefact — the loss is small but not zero.
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Loss (Frames) | Loss Rate | Result |
|---|---|---|---|---|---|---|
| 64 | 446,427,600 | 446,422,979 | 5.000 Gbps | 4,621 | 0% (0.0010%) | Pass |
| 128 | 253,377,600 | 253,374,794 | 5.000 Gbps | 2,806 | 0% (0.0011%) | Pass |
| 256 | 135,868,800 | 135,867,421 | 5.000 Gbps | 1,379 | 0% (0.0010%) | Pass |
| 512 | 70,488,000 | 70,487,322 | 5.000 Gbps | 678 | 0% (0.0010%) | Pass |
| 1024 | 35,918,400 | 35,918,066 | 5.000 Gbps | 334 | 0% (0.0009%) | Pass |
| 1280 | 28,845,600 | 28,845,332 | 5.000 Gbps | 268 | 0% (0.0009%) | Pass |
| 1518 | 24,381,600 | 24,381,372 | 5.000 Gbps | 228 | 0% (0.0009%) | Pass |
| 2048 | 18,133,200 | 18,133,030 | 5.000 Gbps | 170 | 0% (0.0009%) | Pass |
| 4096 | 9,110,400 | 9,110,310 | 5.000 Gbps | 90 | 0% (0.0010%) | Pass |
| 8192 | 4,566,000 | 4,565,950 | 4.999 Gbps | 50 | 0% (0.0011%) | Pass |
| 9000 | 4,156,800 | 4,156,299 | 4.999 Gbps | 501 | 0.01% (0.0121%) | Pass |
Two things are worth drawing out of this table.
The loss is expected, not a fault. The throughput test established that the zero-loss ceiling is 99.609% of line rate. This test forces a fixed 100%, so it deliberately runs ~0.39% above that ceiling. Losing about one frame in 100,000 under that overload is a measure of how gracefully the switch degrades, not of instability. For zero-loss operation, engineer offered load against the 99.61% throughput figure rather than against 100%.
The 9000-byte case is different in kind, not just degree. Its 0.0121% loss is roughly ten times every other frame size, and it is concentrated rather than spread: port P-0-6-3 lost 464 frames at 9000 bytes while the other four ports lost 11 each, and almost all of that sits on the single flow P-0-6-2 → P-0-6-3 (456 frames). Every other frame size distributes its loss evenly across all five ports. If jumbo-frame headroom matters for your design, this is the number to look at.
Per-Port Frame Loss at 100% Rate
Loss is computed per port as Tx − Rx.
Port P-0-6-1
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Rx Rate (L1) | Loss (Frames) |
|---|---|---|---|---|---|
| 64 | 89,285,520 | 89,284,596 | 1000.00 Mbps | 999.99 Mbps | 924 |
| 128 | 50,675,520 | 50,674,954 | 1000.00 Mbps | 999.99 Mbps | 566 |
| 256 | 27,173,760 | 27,173,485 | 999.99 Mbps | 999.98 Mbps | 275 |
| 512 | 14,097,600 | 14,097,463 | 999.99 Mbps | 999.98 Mbps | 137 |
| 1024 | 7,183,680 | 7,183,611 | 999.97 Mbps | 999.96 Mbps | 69 |
| 1280 | 5,769,120 | 5,769,062 | 999.98 Mbps | 999.97 Mbps | 58 |
| 1518 | 4,876,320 | 4,876,270 | 999.97 Mbps | 999.96 Mbps | 50 |
| 2048 | 3,626,640 | 3,626,605 | 999.99 Mbps | 999.98 Mbps | 35 |
| 4096 | 1,822,080 | 1,822,060 | 999.96 Mbps | 999.95 Mbps | 20 |
| 8192 | 913,200 | 913,188 | 999.89 Mbps | 999.88 Mbps | 12 |
| 9000 | 831,360 | 831,349 | 999.85 Mbps | 999.84 Mbps | 11 |
Port P-0-6-2
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Rx Rate (L1) | Loss (Frames) |
|---|---|---|---|---|---|
| 64 | 89,285,520 | 89,284,592 | 1000.00 Mbps | 999.99 Mbps | 928 |
| 128 | 50,675,520 | 50,674,955 | 1000.00 Mbps | 999.99 Mbps | 565 |
| 256 | 27,173,760 | 27,173,481 | 999.99 Mbps | 999.98 Mbps | 279 |
| 512 | 14,097,600 | 14,097,463 | 999.99 Mbps | 999.98 Mbps | 137 |
| 1024 | 7,183,680 | 7,183,612 | 999.97 Mbps | 999.96 Mbps | 68 |
| 1280 | 5,769,120 | 5,769,066 | 999.98 Mbps | 999.97 Mbps | 54 |
| 1518 | 4,876,320 | 4,876,280 | 999.97 Mbps | 999.96 Mbps | 40 |
| 2048 | 3,626,640 | 3,626,606 | 999.99 Mbps | 999.98 Mbps | 34 |
| 4096 | 1,822,080 | 1,822,061 | 999.96 Mbps | 999.95 Mbps | 19 |
| 8192 | 913,200 | 913,189 | 999.89 Mbps | 999.88 Mbps | 11 |
| 9000 | 831,360 | 831,349 | 999.85 Mbps | 999.84 Mbps | 11 |
Port P-0-6-3
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Rx Rate (L1) | Loss (Frames) |
|---|---|---|---|---|---|
| 64 | 89,285,520 | 89,284,591 | 1000.00 Mbps | 999.99 Mbps | 929 |
| 128 | 50,675,520 | 50,674,963 | 1000.00 Mbps | 999.99 Mbps | 557 |
| 256 | 27,173,760 | 27,173,484 | 999.99 Mbps | 999.98 Mbps | 276 |
| 512 | 14,097,600 | 14,097,466 | 999.99 Mbps | 999.98 Mbps | 134 |
| 1024 | 7,183,680 | 7,183,611 | 999.97 Mbps | 999.96 Mbps | 69 |
| 1280 | 5,769,120 | 5,769,068 | 999.98 Mbps | 999.97 Mbps | 52 |
| 1518 | 4,876,320 | 4,876,270 | 999.97 Mbps | 999.96 Mbps | 50 |
| 2048 | 3,626,640 | 3,626,608 | 999.99 Mbps | 999.98 Mbps | 32 |
| 4096 | 1,822,080 | 1,822,061 | 999.96 Mbps | 999.95 Mbps | 19 |
| 8192 | 913,200 | 913,190 | 999.89 Mbps | 999.88 Mbps | 10 |
| 9000 | 831,360 | 830,896 | 999.85 Mbps | 999.29 Mbps | 464 |
Port P-0-6-4
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Rx Rate (L1) | Loss (Frames) |
|---|---|---|---|---|---|
| 64 | 89,285,520 | 89,284,594 | 1000.00 Mbps | 999.99 Mbps | 926 |
| 128 | 50,675,520 | 50,674,954 | 1000.00 Mbps | 999.99 Mbps | 566 |
| 256 | 27,173,760 | 27,173,481 | 999.99 Mbps | 999.98 Mbps | 279 |
| 512 | 14,097,600 | 14,097,463 | 999.99 Mbps | 999.98 Mbps | 137 |
| 1024 | 7,183,680 | 7,183,613 | 999.97 Mbps | 999.96 Mbps | 67 |
| 1280 | 5,769,120 | 5,769,067 | 999.98 Mbps | 999.97 Mbps | 53 |
| 1518 | 4,876,320 | 4,876,269 | 999.97 Mbps | 999.96 Mbps | 51 |
| 2048 | 3,626,640 | 3,626,600 | 999.99 Mbps | 999.97 Mbps | 40 |
| 4096 | 1,822,080 | 1,822,060 | 999.96 Mbps | 999.95 Mbps | 20 |
| 8192 | 913,200 | 913,188 | 999.89 Mbps | 999.88 Mbps | 12 |
| 9000 | 831,360 | 831,349 | 999.85 Mbps | 999.84 Mbps | 11 |
Port P-0-6-5
| Frame Size (bytes) | Tx (Frames) | Rx (Frames) | Tx Rate (L1) | Rx Rate (L1) | Loss (Frames) |
|---|---|---|---|---|---|
| 64 | 89,285,520 | 89,284,598 | 1000.00 Mbps | 999.99 Mbps | 922 |
| 128 | 50,675,520 | 50,674,960 | 1000.00 Mbps | 999.99 Mbps | 560 |
| 256 | 27,173,760 | 27,173,482 | 999.99 Mbps | 999.98 Mbps | 278 |
| 512 | 14,097,600 | 14,097,464 | 999.99 Mbps | 999.98 Mbps | 136 |
| 1024 | 7,183,680 | 7,183,612 | 999.97 Mbps | 999.96 Mbps | 68 |
| 1280 | 5,769,120 | 5,769,064 | 999.98 Mbps | 999.97 Mbps | 56 |
| 1518 | 4,876,320 | 4,876,275 | 999.97 Mbps | 999.96 Mbps | 45 |
| 2048 | 3,626,640 | 3,626,605 | 999.99 Mbps | 999.98 Mbps | 35 |
| 4096 | 1,822,080 | 1,822,061 | 999.96 Mbps | 999.95 Mbps | 19 |
| 8192 | 913,200 | 913,189 | 999.89 Mbps | 999.88 Mbps | 11 |
| 9000 | 831,360 | 831,349 | 999.85 Mbps | 999.84 Mbps | 11 |
Per-port losses sum to a few frames more than the aggregate figure (3–8 frames depending on frame size) because Xena samples the port and aggregate counters at slightly different instants. Both are reproduced here as printed.
Back-to-Back Test
The back-to-back test measures the longest burst the device can forward without loss. Each frame size was tested for 60 seconds across 3 iterations at 50% and 100% rates using a common result scope.
Every one of the 20 flows recorded zero receive loss at both rates and at all eleven frame sizes — received frames matched transmitted frames exactly throughout.
The 100%-rate bursts are markedly shorter than the 50%-rate ones — 124,877,253 frames versus 223,214,280 at 64 bytes. That is the expected shape of a back-to-back curve, not a regression. Back-to-back records the longest burst absorbed with zero loss: at half line rate the switch drains its buffers as fast as they fill, so the loss-free burst runs very long; at full line rate ingress outpaces egress, the buffer eventually saturates, and the loss-free burst necessarily ends sooner.
Results at 50% Rate
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) | Max Offered Rate (fps) | Max Offered Rate (Gbps) | Result |
|---|---|---|---|---|---|
| 64 | 223,214,280 | 14,285,713,920 | 3,638,686 | 2.445 | Pass |
| 128 | 126,689,180 | 16,216,215,040 | 2,064,897 | 2.445 | Pass |
| 256 | 67,934,780 | 17,391,303,680 | 1,107,207 | 2.445 | Pass |
| 512 | 35,244,360 | 18,045,112,320 | 574,494 | 2.445 | Pass |
| 1024 | 17,959,760 | 18,390,794,240 | 292,734 | 2.445 | Pass |
| 1280 | 14,423,060 | 18,461,516,800 | 235,127 | 2.445 | Pass |
| 1518 | 12,191,140 | 18,506,150,520 | 198,720 | 2.445 | Pass |
| 2048 | 9,066,720 | 18,568,642,560 | 147,795 | 2.445 | Pass |
| 4096 | 4,555,380 | 18,658,836,480 | 74,254 | 2.445 | Pass |
| 8192 | 2,283,240 | 18,704,302,080 | 37,213 | 2.445 | Pass |
| 9000 | 2,078,700 | 18,708,300,000 | 33,885 | 2.445 | Pass |
At half line rate the switch absorbed a 223.2-million-frame burst at 64 bytes (14.29 GB) through to 2.08 million frames at 9000 bytes (18.71 GB), without dropping a single frame anywhere. Total burst bytes rise steadily with frame size as per-frame overhead falls.
Per-Port Back-to-Back at 50% Rate
Each of the five ports carried exactly one fifth of the aggregate burst at every frame size. Port P-0-6-1 is shown as representative.
Port P-0-6-1
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) |
|---|---|---|
| 64 | 44,642,856 | 2,857,142,784 |
| 128 | 25,337,836 | 3,243,243,008 |
| 256 | 13,586,956 | 3,478,260,736 |
| 512 | 7,048,872 | 3,609,022,464 |
| 1024 | 3,591,952 | 3,678,158,848 |
| 1280 | 2,884,612 | 3,692,303,360 |
| 1518 | 2,438,228 | 3,701,230,104 |
| 2048 | 1,813,344 | 3,713,728,512 |
| 4096 | 911,076 | 3,731,767,296 |
| 8192 | 456,648 | 3,740,860,416 |
| 9000 | 415,740 | 3,741,660,000 |
Ports P-0-6-2 through P-0-6-5 — identical burst figures at every frame size.
Results at 100% Rate
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) | Max Offered Rate (fps) | Max Offered Rate (Gbps) | Result |
|---|---|---|---|---|---|
| 64 | 124,877,253 | 7,992,144,213 | 7,440,476 | 5.000 | Pass |
| 128 | 62,280,233 | 7,971,869,867 | 4,222,973 | 5.000 | Pass |
| 256 | 41,074,067 | 10,514,961,067 | 2,264,493 | 5.000 | Pass |
| 512 | 24,361,167 | 12,472,917,333 | 1,174,812 | 5.000 | Pass |
| 1024 | 13,100,500 | 13,414,912,000 | 598,659 | 5.000 | Pass |
| 1280 | 10,591,933 | 13,557,674,667 | 480,769 | 5.000 | Pass |
| 1518 | 8,819,060 | 13,387,333,080 | 406,372 | 5.000 | Pass |
| 2048 | 6,529,800 | 13,373,030,400 | 302,224 | 5.000 | Pass |
| 4096 | 2,959,813 | 12,123,395,413 | 151,846 | 5.000 | Pass |
| 8192 | 1,178,120 | 9,651,159,040 | 76,108 | 5.000 | Pass |
| 9000 | 970,160 | 8,731,440,000 | 69,290 | 5.000 | Pass |
At full line rate — the measured offered rate lands within 10 kbit/s of 5 Gbps at every frame size, so this is a genuine 100%-of-line-rate run — the loss-free burst ranges from 124.9 million frames (7.99 GB) at 64 bytes to 970,160 frames (8.73 GB) at 9000 bytes. In byte terms the burst peaks at 13.56 GB around 1280-byte frames rather than rising monotonically: buffer depth measured in bytes is not constant across frame sizes, and large frames retire fewer total bytes before the first loss.
Per-Port Back-to-Back at 100% Rate
Each of the five ports carried an identical burst at every frame size. Port P-0-6-1 is shown as representative.
Port P-0-6-1
| Frame Size (bytes) | Tx Burst (Frames) | Tx Burst (Bytes) |
|---|---|---|
| 64 | 24,975,451 | 1,598,428,843 |
| 128 | 12,456,047 | 1,594,373,973 |
| 256 | 8,214,813 | 2,102,992,213 |
| 512 | 4,872,233 | 2,494,583,467 |
| 1024 | 2,620,100 | 2,682,982,400 |
| 1280 | 2,118,387 | 2,711,534,933 |
| 1518 | 1,763,812 | 2,677,466,616 |
| 2048 | 1,305,960 | 2,674,606,080 |
| 4096 | 591,963 | 2,424,679,083 |
| 8192 | 235,624 | 1,930,231,808 |
| 9000 | 194,032 | 1,746,288,000 |
Ports P-0-6-2 through P-0-6-5 — identical burst figures at every frame size. The aggregate row and the sum of the five port rows differ by up to 2 frames at some sizes, which is rounding in the aggregate row rather than real asymmetry.
About This Test
RFC 2544 — Benchmarking Methodology for Network Interconnect Devices — defines the standard test procedures for measuring throughput, latency, frame loss rate, and back-to-back burst capacity of network devices in a controlled laboratory environment.
Testing was performed with the Xena Networks Xena2544 automated test suite (v2.97), which implements the binary search and timing procedures defined by the RFC. Each test sub-section ran for 60 seconds per frame size across 3 iterations; reported values are averages of those iterations. The full run took 14 hours 18 minutes of wall-clock time across all four test suites.
This run extended the standard RFC 2544 frame-size set beyond 1518 bytes to cover 2048, 4096, 8192 and 9000-byte jumbo frames. The report's own comment records the intent: "Brainboxes PE-515rev4 tested to RFC2544 with frame sizes of 64 to 1518 extended to include frame sizes of 2048,4096,8192,9000 with Xena2544."
The PE-515 is a 5-port embedded Gigabit Ethernet switch in the Brainboxes PE range of Pure Embedded industrial Ethernet switches. See the PE-515 overview for hardware features, the PE-515 specifications for electrical and operating data, the PE-515 design guide for integration guidance, and other RFC 2544 test reports for comparable results across the Brainboxes switch ranges.
