Skip to content

Simple Tuning

Jump to bottom
Kevin Huck edited this page Mar 6, 2023 · 1 revision

The following example will introduce APEX using a simple tuning policy.

APEX is integrated with Active Harmony and other search methods.

Source Code

The following exammple is a simple tuner that will autotune three long values to minimize a target value.

[kehuck1@mahti-login12 apex-tutorial]$ srun apex_exec --apex:tasktree ./build/bin/custom_tuning
  ___  ______ _______   __
 / _ \ | ___ \  ___\ \ / /
/ /_\ \| |_/ / |__  \ V /
|  _  ||  __/|  __| /   \
| | | || |   | |___/ /^\ \
\_| |_/\_|   \____/\/   \/
APEX Version: -62a039a-develop
Built on: 15:44:00 Mar  6 2023 (RelWithDebInfo)
C++ Language Standard version : 201402
GCC Compiler version : 11.2.0
Running custom tuning test
Tuning session 1 handle: 1
Tuning session 2 handle: 2
p1 = 3 p2 = 3 value = 40
x = 76 sv = 0.841471
p1 = 8 p2 = 4 value = 24
x = 26 sv = 0.649541
p1 = 4 p2 = 8 value = 72
x = 51 sv = -0.926185
p1 = 3 p2 = 3 value = 24
x = 26 sv = 0.756695
p1 = 5 p2 = 4 value = 24
x = 1 sv = -0.926185
p1 = 4 p2 = 5 value = 42
x = 26 sv = 0.841471
p1 = 3 p2 = 3 value = 30
x = 38 sv = -0.926185
p1 = 1 p2 = 3 value = 24
x = 26 sv = -0.118492
p1 = 2 p2 = 1 value = 4
x = 13 sv = -0.926185
p1 = 2 p2 = 3 value = 18
x = 26 sv = -0.447492
p1 = 1 p2 = 3 value = 14
x = 32 sv = -0.926185
p1 = 1 p2 = 2 value = 4
x = 26 sv = -0.586176
p1 = 0 p2 = 2 value = 6
x = 19 sv = -0.926185
p1 = 1 p2 = 3 value = -4
x = 26 sv = -0.93817
p1 = 0 p2 = 3 value = 4
x = 13 sv = -0.926185
p1 = 0 p2 = 3 value = -6
x = 26 sv = -0.447492
p1 = 0 p2 = 3 value = -6
x = 19 sv = -0.926185
p1 = 0 p2 = 3 value = -6
x = 13 sv = -0.93817
p1 = 0 p2 = 3 value = -6
x = 19 sv = -0.447492
Tuning has converged for session 1.
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.93817
p1 = 0 p2 = 3 value = -6
x = 26 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.926185
p1 = 0 p2 = 3 value = -6
x = 21 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.991586
p1 = 0 p2 = 3 value = -6
x = 24 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.982642
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
Tuning has converged for session 2.
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
p1 = 0 p2 = 3 value = -6
x = 22 sv = -0.999759
done.
Test passed.

Start Date/Time: 06/03/2023 21:49:09
Elapsed time: 0.0629032 seconds
Total processes detected: 1
HW Threads detected on rank 0: 256
Worker Threads observed on rank 0: 1
Available CPU time on rank 0: 0.0629032 seconds
Available CPU time on all ranks: 0.0629032 seconds

Counter                                              :  #samp |   mean  |  max
--------------------------------------------------------------------------------
                               1 Minute Load average :      1     1.10     1.10
                                      status:Threads :      1     2.00     2.00
                                    status:VmData kB :      1 1.06e+06 1.06e+06
                                     status:VmExe kB :      1    16.00    16.00
                                     status:VmHWM kB :      1 1.01e+04 1.01e+04
                                     status:VmLck kB :      1     0.00     0.00
                                     status:VmLib kB :      1 2.24e+04 2.24e+04
                                     status:VmPTE kB :      1   152.00   152.00
                                    status:VmPeak kB :      1 1.25e+06 1.25e+06
                                     status:VmPin kB :      1     0.00     0.00
                                     status:VmRSS kB :      1 1.01e+04 1.01e+04
                                    status:VmSize kB :      1 1.19e+06 1.19e+06
                                     status:VmStk kB :      1   148.00   148.00
                                    status:VmSwap kB :      1     0.00     0.00
                   status:nonvoluntary_ctxt_switches :      1     1.00     1.00
                      status:voluntary_ctxt_switches :      1    31.00    31.00
--------------------------------------------------------------------------------

CPU Timers                                           : #calls|   mean |   total
--------------------------------------------------------------------------------
                                           APEX MAIN :      1     0.06     0.06
          int apex_preload_main(int, char**, char**) :      1     0.06     0.06
--------------------------------------------------------------------------------


--------------------------------------------------------------------------------
                                        Total timers : 1
Writing: .//apex_tasktree.csv

APEX tutorial, © Copyright 2023, University of Oregon. For more information on APEX, see https://github.com/UO-OACISS/apex

Home

Programming Model Examples

  1. C POSIX threads
  2. Standard CPP threads
  3. OpenMP Examples
  4. OpenMP Target Offload
  5. MPI and OpenMP Example
  6. HIP Example
  7. CUDA Example
  8. Kokkos Example

APEX Features

  1. General Features
  2. Data Collection Features
  3. Device Features
  4. Utilities
  5. Post Processing Utilities

Policy Examples

  1. Simple Tuning
  2. Mixed Type Tuning
Clone this wiki locally