For a discussion on stack overflow

perf counters for remove, 10k ints

Different times for different alignments: best is padding is 6 bytes, worth when it's 50 bytes
===============================================================================================================
sudo perf stat -e LSD.UOPS,UOPS_ISSUED.ANY,branches,branch-misses  ./src/bench_runnable2/remove "--benchmark_filter=.*size:10000/type:int/algorithm:std::remove/percentage:50/padding:50$"
2020-04-26 00:59:16
Running ./src/bench_runnable2/remove
Run on (8 X 4900 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x8)
  L1 Instruction 32 KiB (x8)
  L2 Unified 256 KiB (x8)
  L3 Unified 12288 KiB (x1)
Load Average: 0.21, 0.12, 0.08
-----------------------------------------------------------------------------------------------------------------------------
Benchmark                                                                                   Time             CPU   Iterations
-----------------------------------------------------------------------------------------------------------------------------
/name:remove 0s/size:10000/type:int/algorithm:std::remove/percentage:50/padding:50       7211 ns         7211 ns        97153

 Performance counter stats for './src/bench_runnable2/remove --benchmark_filter=.*size:10000/type:int/algorithm:std::remove/percentage:50/padding:50$':

                 0      LSD.UOPS                                                    
    69,442,388,854      UOPS_ISSUED.ANY                                             
    11,826,143,943      branches                                                    
       127,890,842      branch-misses             #    1.08% of all branches        

       5.302266596 seconds time elapsed

       5.268550000 seconds user
       0.016013000 seconds sys


sudo perf stat -e LSD.UOPS,UOPS_ISSUED.ANY,branches,branch-misses  ./src/bench_runnable2/remove "--benchmark_filter=.*size:10000/type:int/algorithm:std::remove/percentage:50/padding:6$"
2020-04-26 00:59:29
Running ./src/bench_runnable2/remove
Run on (8 X 4900 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x8)
  L1 Instruction 32 KiB (x8)
  L2 Unified 256 KiB (x8)
  L3 Unified 12288 KiB (x1)
Load Average: 0.24, 0.13, 0.08
----------------------------------------------------------------------------------------------------------------------------
Benchmark                                                                                  Time             CPU   Iterations
----------------------------------------------------------------------------------------------------------------------------
/name:remove 0s/size:10000/type:int/algorithm:std::remove/percentage:50/padding:6       1460 ns         1460 ns       475866

 Performance counter stats for './src/bench_runnable2/remove --benchmark_filter=.*size:10000/type:int/algorithm:std::remove/percentage:50/padding:6$':

                 0      LSD.UOPS                                                    
    74,238,458,573      UOPS_ISSUED.ANY                                             
    14,824,364,356      branches                                                    
        19,807,323      branch-misses             #    0.13% of all branches        

       5.371986435 seconds time elapsed

       5.333160000 seconds user
       0.020004000 seconds sys