14 | | === Hardware and Software === |
15 | | |
16 | | * Netgear ProSafe 5-port Gigabit Desktop Switch (GS105) |
17 | | * PC Hardware |
18 | | * Dell Optiplex 790 |
19 | | * Intel Core i7-2600 3.4 GHz |
20 | | * 16 GB RAM |
21 | | * Window 7 Professional SP1 64-bit |
22 | | * Intel 82579LM Gigabit Network Connection |
23 | | * Configured to support Jumbo Packets of up to 9014 Bytes |
24 | | * PC Software |
25 | | * Matlab 7.12.0 (R2011a) |
26 | | * Microsoft Visual C++ 2010 Express (compiler for MEX) |
27 | | * Microsoft Windows 7 SDK |
28 | | [[BR]] |
29 | | === Procedure === |
30 | | |
31 | | * Download Reference bit stream w/ iMPACT |
32 | | * Open MATLAB |
33 | | * Set current folder within MATLAB to the M_Code_Reference folder of the appropriate WARPLab release |
34 | | * Run the following commands |
35 | | * wl_setup |
36 | | * NOTE: This allows us to configure the Transport type and the Ethernet Packet Size |
37 | | * nodes = wl_initNodes(1) |
38 | | * results = wl_benchmark( nodes, 'write_iq', 10000, 'max' ); |
39 | | * results = wl_benchmark( nodes, 'read_iq', 10000, 'max' ); |
40 | | [[BR]] |
41 | | === Results === |
42 | | |
43 | | ''' WARP v3 Hardware ''' |
44 | | |
45 | | Over 10,000 trials using 32768 samples per trial, we have the following Read IQ and Write IQ performance of the different Transports: |
46 | | |
47 | | [[Image(WARP_v3_Ethernet_Performance.png)]] |
48 | | |
49 | | {{{ |
50 | | ------------------------------------------------------------------------------------------------------------------------------------ |
51 | | | ID | WLVER | Serial # | Transport | Pkt Size (B) | Num Samples | Num Trials | Avg Write IQ per sec | Avg Read IQ per sec | |
52 | | ------------------------------------------------------------------------------------------------------------------------------------ |
53 | | | 0 | 7.3.0 | W3-a-00027 | WL Mex UDP | 1464 | 32768 | 10000 | 210.4994 | 232.6682 | |
54 | | | 0 | 7.3.0 | W3-a-00027 | Java UDP | 1464 | 32768 | 10000 | 17.0187 | 16.9429 | |
55 | | | 0 | 7.1.0 | W3-a-00027 | Java UDP | 1464 | 32768 | 10000 | 16.9390 | 16.6640 | |
56 | | ------------------------------------------------------------------------------------------------------------------------------------ |
57 | | | 0 | 7.3.0 | W3-a-00027 | WL Mex UDP | 8960 | 32768 | 10000 | 445.8885 | 466.6237 | |
58 | | | 0 | 7.3.0 | W3-a-00027 | Java UDP | 8960 | 32768 | 10000 | 82.0430 | 71.2319 | |
59 | | | 0 | 7.1.0 | W3-a-00027 | Java UDP | 8960 | 32768 | 10000 | 73.1989 | 44.6655 | |
60 | | ------------------------------------------------------------------------------------------------------------------------------------ |
61 | | }}} |
62 | | |
63 | | |
64 | | |
65 | | |
66 | | ''' WARP v2 Hardware ''' |
67 | | |
68 | | Over 10,000 trials using 16384 samples per trial, we have the following Read IQ and Write IQ performance of the different Transports: |
69 | | |
70 | | [[Image(WARP_v2_Ethernet_Performance.png)]] |
71 | | |
72 | | {{{ |
73 | | ------------------------------------------------------------------------------------------------------------------------------------ |
74 | | | ID | WLVER | Serial # | Transport | Pkt Size (B) | Num Samples | Num Trials | Avg Write IQ per sec | Avg Read IQ per sec | |
75 | | ------------------------------------------------------------------------------------------------------------------------------------ |
76 | | | 0 | 7.3.0 | N/A | WL Mex UDP | 1464 | 16384 | 10000 | 128.1127 | 133.0076 | |
77 | | | 0 | 7.3.0 | N/A | Java UDP | 1464 | 16384 | 10000 | 32.9602 | 31.1658 | |
78 | | | 0 | 7.1.0 | N/A | Java UDP | 1464 | 16384 | 10000 | 32.5807 | 22.0276 | |
79 | | ------------------------------------------------------------------------------------------------------------------------------------ |
80 | | }}} |
81 | | |
| 14 | - Ethernet Performance Benchmarks |
| 15 | - [wiki:WARPLab/Benchmarks/WARPLAB_7_3_0 WARPLab 7.3.0] |
| 16 | - WARPLab 7.4.0 benchmarks are the same as WARPLab 7.3.0 |
| 17 | - [wiki:WARPLab/Benchmarks/WARPLAB_7_5_0 WARPLab 7.5.0] |