This is a material for overview of ppOpen-AT/Static for ppOpen-APPL/FDM ver. 0.2.0, which is numerical simulation software of a seismic wave analysis with function of automatic performance tuning (AT). Project of ppOpen-HPC is developing and supporting for this software. The effect of AT is shown with respect to several recent computer environments, such as multi-core (Ivy Bridge) and many-core (Xeon Phi).

1. Overview of ppOpen-AT/Static
for ppOpen-APPL/FDM ver. 0.2.0
Takahiro Katagiri
(Information Technology Center,
The University of Tokyo)
1
As of July 30th, 2014

2. ppOpen‐AT System
ppOpen‐APPL /*
ppOpen‐AT
Directives
User
KnowledgeLibrary
Developer
① Before
Release‐time
Candidate
1
Candidate
2
Candidate
3
Candidate
nppOpen‐AT
Auto‐Tuner
ppOpen‐APPL / *
Automatic
Code
Generation②
:Target
Computers
Execution Time④
Library User
③
Library Call
Selection
⑤
⑥
Auto‐tuned
Kernel
Execution
Run‐
time

3. A Scenario to Software Developers for
ppOpen-AT
3
Executable Code with
Optimization Candidates
and AT Function
Software
Developer
Description of AT by Using
ppOpen-AT
Invocate dedicated
Preprocessor
Program with AT
Functions
Optimization
that cannot be
established by
compilers
■Automatic Generated
Functions
Optimization
Candidates
Performance Monitor
Parameter Search
Performance Modeling
#pragma oat install unroll (i,j,k) region start
#pragma oat varied (i,j,k) from 1 to 8
for(i = 0 ; i < n ; i++){
for(j = 0 ; j < n ; j++){
for(k = 0 ; k < n ; k++){
A[i][j]=A[i][j]+B[i][k]*C[k][j]; }}}
#pragma oat install unroll (i,j,k) region end
Description By Software Developer
Optimizations for Source Codes,
Computer Resource, Power Consumption

4. Target Application
 Seism_3D: Simulation for seismic wave analysis.
 Developed by Professor Furumura at the
University of Tokyo.
◦ The code is re-constructed as
ppOpen-APPL/FDM.
◦ version 0.2.0
 OMP parallelization is implemented.
 Finite Differential Method (FDM)
 3D simulation
◦ 3D arrays are allocated.
 Data type: Single Precision (real*4)
4

5. Target Computation Kernels
in Current Implementation
1. Kernel update_stress
◦ 8 Kinds of Candidates with Loop fusion and Loop Split.
2. Kernel update_vel
◦ 5 Kinds of Candidates with Loop Fusion and Loop Split, and
reordering of statements.
3 Kinds of Candidates with Loop Fusion and Loop Split.
3. Kernel update_stress_sponge
4. Kernel update_vel_sponge
5. Kernel ppohFDM_pdiffx3_p4
6. Kernel ppohFDM_pdiffx3_m4
7. Kernel ppohFDM_pdiffy3_p4
8. Kernel ppohFDM_pdiffy3_m4
9. Kernel ppohFDM_pdiffz3_p4
10. Kernel ppohFDM_pdiffz3_m4
5

6. Additional Kernels in Next Release
(Planned in Nov. 2014)
 3 Kinds of Candidates with Loop Fusion
and Loop Split for data packing and
unpacking.
11. Kernel ppohFDM_ps_pack
12. Kernel ppohFDM_ps_unpack
13. Kernel ppohFDM_pv_pack
14. Kernel ppohFDM_pv_unpack
6

7. An Example of Seism_3D Simulation
 West part earthquake inTottori prefecture in Japan
at year 2000. ([1], pp.14)
 The region of 820km x 410km x 128 km is discretized with 0.4km.
 NX x NY x NZ = 2050 x 1025 x 320 ≒ 6.4 : 3.2 : 1.
[1] T. Furumura,“Large-scale Parallel FDM Simulation for Seismic Waves and Strong Shaking”, Supercomputing News,
Information Technology Center,The University of Tokyo,Vol.11, Special Edition 1, 2009. In Japanese.
Figure : Seismic wave translations in west part earthquake in Tottori prefecture in Japan. (a)
Measured waves; (b) Simulation results; (Reference : [1] in pp.13)

8. Maximum AT Effect (Xeon Phi, 8 Nodes)
558
200
171
30 20 51
Speedup [%]

9. Maximum AT Effect (Ivy Bridge, 8 Nodes)
48
42
67
29
47
23
Speedup [%]

10. Maximum AT Effect (FX10, 8 Nodes)
10 9 7
37
46
2
Speedup [%]