Help is available by moving the cursor above any symbol or by checking MAQAO website.
- r0: g3e_1_thread
- r1: grace_1_thread
- r2: g3e_64_thread
- r3: grace_64_thread
Metric | r0 | r1 | r2 | r3 |
---|
Total Time (s) | 460.82 | 364.77 | 25.45 | 10.91 |
Profiled Time (s) | 460.51 | 364.56 | 25.04 | 10.53 |
Time in analyzed loops (%) | 99.4 | 99.6 | 91.1 | 85.4 |
Time in analyzed innermost loops (%) | 92.0 | 89.5 | 81.8 | 78.7 |
Time in user code (%) | 99.4 | 99.6 | 91.2 | 85.6 |
Compilation Options Score (%) | 75.0 | 75.0 | 75.0 | 75.0 |
Array Access Efficiency (%) | Not Available | Not Available | Not Available | Not Available |
|
Potential Speedups |
Perfect Flow Complexity | 1.00 | 1.00 | 1.00 | 1.00 |
Perfect OpenMP + MPI + Pthread | 1.00 | 1.00 | 1.08 | 1.15 |
Perfect OpenMP + MPI + Pthread + Perfect Load Distribution | 1.00 | 1.00 | 1.21 | 1.35 |
No Scalar Integer | Potential Speedup | 1.05 | 1.06 | 1.06 | 1.04 |
Nb Loops to get 80% | 1 | 1 | 1 | 1 |
FP Vectorised | Potential Speedup | 1.00 | 1.00 | 1.00 | 1.00 |
Nb Loops to get 80% | 1 | 1 | 1 | 1 |
Fully Vectorised | Potential Speedup | 1.03 | 1.20 | 1.04 | 1.17 |
Nb Loops to get 80% | 1 | 2 | 1 | 2 |
Only FP Arithmetic | Potential Speedup | 1.06 | 1.46 | 1.07 | 1.35 |
Nb Loops to get 80% | 1 | 2 | 1 | 2 |
Source Object | Issue |
▼spmxv.exe– | |
▼ooo_cmdline.h– | |
○ | -funroll-loops is missing. |
▼ooo_cmdline.cpp– | |
○ | -funroll-loops is missing. |
▼main.cpp– | |
○ | -funroll-loops is missing. |
Source Object | Issue |
▼spmxv.exe– | |
▼ooo_cmdline.h– | |
○ | -funroll-loops is missing. |
▼ooo_cmdline.cpp– | |
○ | -funroll-loops is missing. |
▼main.cpp– | |
○ | -funroll-loops is missing. |
Source Object | Issue |
▼spmxv.exe– | |
▼ooo_cmdline.h– | |
○ | -funroll-loops is missing. |
▼ooo_cmdline.cpp– | |
○ | -funroll-loops is missing. |
▼main.cpp– | |
○ | -funroll-loops is missing. |
Source Object | Issue |
▼spmxv.exe– | |
▼ooo_cmdline.h– | |
○ | -funroll-loops is missing. |
▼ooo_cmdline.cpp– | |
○ | -funroll-loops is missing. |
▼main.cpp– | |
○ | -funroll-loops is missing. |
| r0 | r1 | r2 | r3 |
Experiment Name | | | | |
Application | ./spmxv.exe | same as r0 | same as r0 | same as r0 |
Timestamp | 2024-07-04 14:55:05 | 2024-07-04 17:21:44 | same as r0 | same as r1 |
Experiment Type | Sequential | same as r0 | OpenMP; | same as r2 |
Machine | ip-172-31-42-13 | p11-grace01.cs.it4i.cz | same as r0 | same as r1 |
Architecture | aarch64 | | same as r0 | |
Micro Architecture | ARM_NEOVERSE_V1 | | same as r0 | |
Model Name | | | | |
Cache Size | | | | |
Number of Cores | | | | |
Maximal Frequency | 0 GHz | 3.42 GHz | same as r0 | same as r1 |
OS Version | Linux 6.5.0-1021-aws #21~22.04.1-Ubuntu SMP Fri May 10 21:19:37 UTC 2024 | Linux 5.14.0-362.18.1.el9_3.aarch64 #1 SMP PREEMPT_DYNAMIC Thu Jan 25 07:56:00 UTC 2024 | same as r0 | same as r1 |
Architecture used during static analysis | aarch64 | same as r0 | same as r0 | same as r0 |
Micro Architecture used during static analysis | ARM_NEOVERSE_V1 | same as r0 | same as r0 | same as r0 |
Compilation Options |
spmxv.exe: GNU C++17 12.2.0 -mlittle-endian -mabi=lp64 -mcpu=zeus+crypto+sha3+sm4+nodotprod+noprofile -g -Ofast -fopenmp | spmxv.exe: GNU C++17 12.2.0 -mlittle-endian -mabi=lp64 -mcpu=demeter+crypto+rcpc+sve2-sm4+sve2-aes+sve2-sha3+nodotprod+noprofile+norng+nomemtag+nopredres+nopauth -g -Ofast -fopenmp | same as r0 | same as r1 |
Number of processes observed | 1 | same as r0 | same as r0 | same as r0 |
Number of threads observed | 1 | same as r0 | 64 | same as r2 |
Frequency Driver | NA | cppc_cpufreq | same as r0 | same as r1 |
Frequency Governor | NA | performance | same as r0 | same as r1 |
Huge Pages | madvise | always | same as r0 | same as r1 |
Hyperthreading | off | same as r0 | same as r0 | same as r0 |
Number of sockets | 1 | 2 | same as r0 | same as r1 |
Number of cores per socket | 64 | 72 | same as r0 | same as r1 |
MAQAO version | 2.20.4 | same as r0 | same as r0 | same as r0 |
MAQAO build | fe14d3adfb3256b508e58d35c4c7bad8c581967c::20240703-182948 | same as r0 | same as r0 | same as r0 |
Comments | | same as r0 | same as r0 | same as r0 |