Track Brian Kent's issue.
Comparing CASA-5 and CASA-6 across the two different CPUs available for batch processing in NM and CV shows that the newer CPUs (E5-2640v3) run a simple calibration job (6.7GB) about 1.25 times faster than the old CPUs (E5-2670) with CASA-6 performing slower in every case.
RHEL7 (Times are in minutes with hh:mm in paranthises)
| CASA | nmpost051 (E5-2640v3) | cvpost020 (E5-2640v3) | nmpost038 (E5-2670) | cvpost003 (E5-2670) |
|---|---|---|---|---|
| 5 | 114 (1:54), 117 (1:57) | 110 (1:50), 111 (1:51) | 144 (2:22), 143 (2:23) | 140 (2:20), 141 (2:21) |
| 6 | 156 (2:36), 164 (2:44) | 156 (2:36), 158 (2:38) | 200 (3:20), 201 (3:21) | 197 (3:17), 199 (3:19) |
Running just hifv_importdata() on a larger data set (350GB) shows that nmpost nodes run about 2% to 10% faster than similar cvpost nodes with CASA-6 performing slower in every case.
RHEL7 (Times are in minutes with hh:mm in paranthises)
| CASA | nmpost051 (E5-2640v3) | cvpost020 (E5-2640v3) | nmpost038 (E5-2670) | cvpost003 (E5-2670) |
|---|---|---|---|---|
| 5 | 192 (3:12) | 196 (3:16) | 239 (3:59) | 251 (4:11) |
| 6 | 328 (5:28) | 364 (6:04), 378 (6:18) | 411 (6:51), 427 (7:07) | 453 (7:33) |
Running both hifv_importdata() and hifv_hanning().
RHEL6
RHEL7 (Times are in minutes with hh:mm in paranthises)
| CASA | nmpost051 (E5-2640v3) | cvpost020 (E5-2640v3) | nmpost038 (E5-2670) | cvpost003 (E5-2670) |
|---|---|---|---|---|
| 5 | 343 (5:43) | 413 (6:53) | ||
| 6 |
Running entire pipeline with -n 8 (times are in minutes)
RHEL6
RHEL7 (Times are in minutes with hh:mm in paranthises)
| CASA | nmpost051 (E5-2640v3) | cvpost020 (E5-2640v3) | nmpost038 (E5-2670) | cvpost003 (E5-2670) |
|---|---|---|---|---|
| 5 | 8336* | 5850* | 12701* | |
| 6 |
"*" Means it completed with pbcor errors