Link Copied
Did you try another sample binary, like matrix sample.
sdp@d404e64b801a:~/workspace$ aps ~/intel/vtune/samples/matrix/matrix
vtune: Warning: Memory bandwidth collection requires the sampling driver to be enabled on the system. Disable the "Analyze memory bandwidth" knob to proceed with the analysis or install the sampling driver on the system. See the Sampling Drivers help topic for more details. Note that memory bandwidth collection is not possible if you are profiling inside a virtualized environment.
vtune: Warning: The following events cannot be collected: FRONTEND_RETIRED.LATENCY_GE_4
Addr of buf1 = 0x7faff6dff010
Offs of buf1 = 0x7faff6dff180
Addr of buf2 = 0x7fadf6dfe010
Offs of buf2 = 0x7fadf6dfe1c0
Addr of buf3 = 0x7fabf6dfd010
Offs of buf3 = 0x7fabf6dfd100
Addr of buf4 = 0x7fa9f6dfc010
Offs of buf4 = 0x7fa9f6dfc140
Threads #: 16 requested OpenMP threads
Matrix size: 32768
Using multiply kernel: multiply5
Freq = 3.399998 GHz
Execution time = 12.604 seconds
| Summary information
|--------------------------------------------------------------------
Application : matrix
Report creation date : 2025-05-07 23:07:09
OpenMP threads number per Process: 16
HW Platform : Intel(R) Xeon(R) Processor code named Graniterapids
Frequency : 2.50 GHz
Logical core count per node : 192
Collector type : Driverless Perf system-wide counting
Used statistics : /home/sdp/workspace/aps_result_20250507/d404e64b801a.jf.intel.com
|
| Your application might underutilize the available logical CPU cores
| because of insufficient parallel work, blocking on synchronization, or too much I/O. Perform function or source line-level profiling with tools like Intel(R) VTune(TM) Profiler to discover why the CPU is underutilized.
|
Elapsed Time: 21.69 s
SP GFLOPS: 0.11
DP GFLOPS: 0.00
Average CPU Frequency: 2.69 GHz
IPC Rate: 1.43
Serial Time: 4.65 s 21.43% of Elapsed Time
| The Serial Time of your application is significant. It directly impacts
| application Elapsed Time and scalability.Explore options for parallelization
| with Intel(R) Advisor or algorithm or microarchitecture tuning of the serial
| code with Intel(R) VTune(TM) Profiler.
OpenMP Imbalance: 0.00 s 0.00% of Elapsed Time
Physical Core Utilization: 15.90%
| The metric is below 80% threshold, which may signal a poor physical CPU cores
| utilization caused by: load imbalance, threading runtime overhead, contended
| synchronization, insufficient parallelism, incorrect affinity that utilizes
| logical cores instead of physical cores. Perform threading analysis with tools
| like Intel(R) VTune(TM) Profiler to discover why physical cores are
| underutilized.
Average Physical Core Utilization: 15.29 out of 96 Physical Cores
Memory Stalls: 28.50% of Pipeline Slots
| The metric value can indicate that a significant fraction of execution
| pipeline slots could be stalled due to demand memory load and stores. See the
| second level metrics to define if the application is cache- or DRAM-bound and
| the NUMA efficiency. Use Intel(R) VTune(TM) Profiler Memory Access analysis to
| review a detailed metric breakdown by memory hierarchy, memory bandwidth
| information, and correlation by memory objects.
Cache Stalls: 28.00% of Cycles
| A significant proportion of cycles are spent on data fetches from cache. Use
| Intel(R) VTune(TM) Profiler Memory Access analysis to see if accesses to L2 or
| L3 cache are problematic and consider applying the same performance tuning as
| you would for a cache-missing workload. This may include reducing the data
| working set size, improving data access locality, blocking or partitioning the
| working set to fit in the lower cache levels, or exploiting hardware
| prefetchers.
DRAM Stalls: 1.60% of Cycles
Average DRAM Bandwidth: N/A
| Data for this metric is not collected since it requires system-wide
| performance monitoring. Make sure the sampling driver is properly installed on
| your system:
| https://www.intel.com/content/www/us/en/docs/vtune-profiler/user-guide/current/sep-driver.html.
| Otherwise, enable a driverless Perf-based sampling collection by setting the
| /proc/sys/kernel/perf_even_paranoid value to 0 or less.
Vectorization: 100.00%
Instruction Mix:
Memory Footprint:
Resident: 10525.00 MB
Virtual: 34949.00 MB
