{"id":922,"date":"2020-04-02T02:30:49","date_gmt":"2020-04-01T17:30:49","guid":{"rendered":"https:\/\/wp.study3.biz\/?p=922"},"modified":"2020-04-02T02:44:40","modified_gmt":"2020-04-01T17:44:40","slug":"oracle-linux-server-8-1-titan-rtx-x2-rtx2080tix2-cuda10-2-e5-2650-v4x2-namd-2-12-171025-stmv-virus-benchmark-%e3%82%92%e5%8b%95%e4%bd%9c%e3%81%95%e3%81%9b%e3%81%a6%e3%81%bf%e3%81%9f1066628-atom","status":"publish","type":"post","link":"https:\/\/wp.study3.biz\/?p=922","title":{"rendered":"Oracle Linux Server 8.1 TITAN RTX x2 RTX2080Tix2 CUDA10..2 E5 2650 V4x2 namd 2.12-171025 STMV (virus) benchmark \u3092\u52d5\u4f5c\u3055\u305b\u3066\u307f\u305f(1,066,628 atoms, periodic, PME) 0.349202 days\/ns"},"content":{"rendered":"

\u53c2\u8003\u30b5\u30a4\u30c8\uff11<\/a>\u30b5\u30a4\u30c8\uff12<\/a><\/p>\n

[chibi@oracle8 ~]$ cat \/etc\/os-release
\nNAME=”Oracle Linux Server”
\nVERSION=”8.1″
\nID=”ol”
\nID_LIKE=”fedora”
\nVARIANT=”Server”
\nVARIANT_ID=”server”
\nVERSION_ID=”8.1″
\nPLATFORM_ID=”platform:el8″
\nPRETTY_NAME=”Oracle Linux Server 8.1″
\nANSI_COLOR=”0;31″
\nCPE_NAME=”cpe:\/o:oracle:linux:8:1:server”
\nHOME_URL=”https:\/\/linux.oracle.com\/”
\nBUG_REPORT_URL=”https:\/\/bugzilla.oracle.com\/”<\/p>\n

ORACLE_BUGZILLA_PRODUCT=”Oracle Linux 8″
\nORACLE_BUGZILLA_PRODUCT_VERSION=8.1
\nORACLE_SUPPORT_PRODUCT=”Oracle Linux”
\nORACLE_SUPPORT_PRODUCT_VERSION=8.1
\n[chibi@oracle8 ~]$ cat \/etc\/redhat-release
\nRed Hat Enterprise Linux release 8.1 (Ootpa)
\n[chibi@oracle8 ~]$ nvcc -V
\nnvcc: NVIDIA (R) Cuda compiler driver
\nCopyright (c) 2005-2019 NVIDIA Corporation
\nBuilt on Wed_Oct_23_19:24:38_PDT_2019
\nCuda compilation tools, release 10.2, V10.2.89
\n[chibi@oracle8 ~]$ sudo nvidia-docker run -it –rm nvcr.io\/hpc\/namd:2.12-171025 \/opt\/namd\/namd-multicore-memopt +p40 +setcpuaffinity +idlepoll \/workspace\/examples\/stmv\/stmv_pmecuda.namd
\n[sudo] chibi \u306e\u30d1\u30b9\u30ef\u30fc\u30c9:
\nUnable to find image ‘nvcr.io\/hpc\/namd:2.12-171025’ locally
\n2.12-171025: Pulling from hpc\/namd
\nf6fa9a861b90: Pull complete
\n2d93875543ec: Pull complete
\n407421ef3e7e: Pull complete
\nea9ffec33008: Pull complete
\nc695ce24f66e: Pull complete
\ncb6e6f26f62f: Pull complete
\n4ca5cacd5888: Pull complete
\n127359e380ae: Pull complete
\n09f52fb90f32: Pull complete
\nc8b4fccff7c3: Pull complete
\na898f5b12168: Pull complete
\n0c9fa151e12b: Pull complete
\ne9f9e8f970e4: Pull complete
\n6cb19c6e7375: Pull complete
\n76db9e80d16b: Pull complete
\nDigest: sha256:c9184f9b071f2197f20a0064ed47f9dc8deac9f007e03d384ecbaad28a754124
\nStatus: Downloaded newer image for nvcr.io\/hpc\/namd:2.12-171025
\nCharm++: standalone mode (not using charmrun)
\nCharm++> Running in Multicore mode: 40 threads
\nCharm++> Using recursive bisection (scheme 3) for topology aware partitions
\nConverse\/Charm++ Commit ID: v6.8.2
\nWarning> Randomization of virtual memory (ASLR) is turned on in the kernel, thread migration may not work! Run ‘echo 0 > \/proc\/sys\/kernel\/randomize_va_space’ as root to disable it, or try running with ‘+isomalloc_sync’.
\nCharmLB> Load balancer assumes all CPUs are same.
\nCharm++> cpu affinity enabled.
\nCharm++> Running on 1 unique compute nodes (48-way SMP).
\nCharm++> cpu topology info is gathered in 0.019 seconds.
\nInfo: Built with CUDA version 9000
\nDid not find +devices i,j,k,… argument, using all
\nPe 0 physical rank 0 will use CUDA device of pe 8
\nPe 36 physical rank 36 will use CUDA device of pe 32
\nPe 33 physical rank 33 will use CUDA device of pe 32
\nPe 9 physical rank 9 will use CUDA device of pe 8
\nPe 18 physical rank 18 will use CUDA device of pe 16
\nPe 39 physical rank 39 will use CUDA device of pe 32
\nPe 22 physical rank 22 will use CUDA device of pe 24
\nPe 15 physical rank 15 will use CUDA device of pe 16
\nPe 20 physical rank 20 will use CUDA device of pe 24
\nPe 25 physical rank 25 will use CUDA device of pe 24
\nPe 7 physical rank 7 will use CUDA device of pe 8
\nPe 3 physical rank 3 will use CUDA device of pe 8
\nPe 11 physical rank 11 will use CUDA device of pe 16
\nPe 27 physical rank 27 will use CUDA device of pe 24
\nPe 12 physical rank 12 will use CUDA device of pe 16
\nPe 30 physical rank 30 will use CUDA device of pe 32
\nPe 6 physical rank 6 will use CUDA device of pe 8
\nPe 17 physical rank 17 will use CUDA device of pe 16
\nPe 13 physical rank 13 will use CUDA device of pe 16
\nPe 31 physical rank 31 will use CUDA device of pe 32
\nPe 37 physical rank 37 will use CUDA device of pe 32
\nPe 1 physical rank 1 will use CUDA device of pe 8
\nPe 35 physical rank 35 will use CUDA device of pe 32
\nPe 4 physical rank 4 will use CUDA device of pe 8
\nPe 28 physical rank 28 will use CUDA device of pe 24
\nPe 38 physical rank 38 will use CUDA device of pe 32
\nPe 21 physical rank 21 will use CUDA device of pe 24
\nPe 14 physical rank 14 will use CUDA device of pe 16
\nPe 5 physical rank 5 will use CUDA device of pe 8
\nPe 19 physical rank 19 will use CUDA device of pe 16
\nPe 26 physical rank 26 will use CUDA device of pe 24
\nPe 2 physical rank 2 will use CUDA device of pe 8
\nPe 34 physical rank 34 will use CUDA device of pe 32
\nPe 23 physical rank 23 will use CUDA device of pe 24
\nPe 10 physical rank 10 will use CUDA device of pe 16
\nPe 29 physical rank 29 will use CUDA device of pe 24
\nPe 32 physical rank 32 binding to CUDA device 3 on fef16d337343: ‘GeForce RTX 2080 Ti’ Mem: 11019MB Rev: 7.5
\nPe 24 physical rank 24 binding to CUDA device 2 on fef16d337343: ‘GeForce RTX 2080 Ti’ Mem: 11019MB Rev: 7.5
\nPe 16 physical rank 16 binding to CUDA device 1 on fef16d337343: ‘TITAN RTX’ Mem: 24220MB Rev: 7.5
\nPe 8 physical rank 8 binding to CUDA device 0 on fef16d337343: ‘TITAN RTX’ Mem: 24220MB Rev: 7.5
\nInfo: NAMD 2.12 for Linux-x86_64-multicore-CUDA-memopt
\nWarning:
\nWarning: *** EXPERIMENTAL MEMORY OPTIMIZED VERSION ***
\nWarning:
\nInfo:
\nInfo: Please visit http:\/\/www.ks.uiuc.edu\/Research\/namd\/
\nInfo: for updates, documentation, and support information.
\nInfo:
\nInfo: Please cite Phillips et al., J. Comp. Chem. 26:1781-1802 (2005)
\nInfo: in all publications reporting results obtained with NAMD.
\nInfo:
\nInfo: Based on Charm++\/Converse 60800 for multicore-linux64-gcc
\nInfo: Built Tue Nov 21 02:03:10 UTC 2017 by on a02d2dbfe66b
\nInfo: 1 NAMD 2.12 Linux-x86_64-multicore-CUDA-memopt 40 fef16d337343 root
\nInfo: Running on 40 processors, 1 nodes, 1 physical nodes.
\nInfo: CPU topology information available.
\nInfo: Charm++\/Converse parallel runtime startup completed at 1.28777 s
\nCkLoopLib is used in SMP with a simple dynamic scheduling (converse-level notification) but not using node-level queue
\nInfo: 38.5625 MB of memory in use based on \/proc\/self\/stat
\nInfo: Configuration file is \/workspace\/examples\/stmv\/stmv_pmecuda.namd
\nInfo: Changed directory to \/workspace\/examples\/stmv
\nTCL: Suspending until startup complete.
\nInfo: SIMULATION PARAMETERS:
\nInfo: TIMESTEP 1
\nInfo: NUMBER OF STEPS 800
\nInfo: STEPS PER CYCLE 20
\nInfo: PERIODIC CELL BASIS 1 216.832 0 0
\nInfo: PERIODIC CELL BASIS 2 0 216.832 0
\nInfo: PERIODIC CELL BASIS 3 0 0 216.832
\nInfo: PERIODIC CELL CENTER 0 0 0
\nInfo: LOAD BALANCER Hybrid
\nInfo: LOAD BALANCING STRATEGY New Load Balancers — DEFAULT
\nInfo: LDB PERIOD 4000 steps
\nInfo: FIRST LDB TIMESTEP 100
\nInfo: HYBRIDLB GROUP SIZE 512
\nInfo: LAST LDB TIMESTEP -1
\nInfo: LDB BACKGROUND SCALING 1
\nInfo: HOM BACKGROUND SCALING 1
\nInfo: PME BACKGROUND SCALING 1
\nInfo: MAX SELF PARTITIONS 1
\nInfo: MAX PAIR PARTITIONS 1
\nInfo: SELF PARTITION ATOMS 154
\nInfo: SELF2 PARTITION ATOMS 154
\nInfo: PAIR PARTITION ATOMS 318
\nInfo: PAIR2 PARTITION ATOMS 637
\nInfo: MIN ATOMS PER PATCH 40
\nInfo: INITIAL TEMPERATURE 298
\nInfo: CENTER OF MASS MOVING INITIALLY? NO
\nInfo: DIELECTRIC 1
\nInfo: EXCLUDE SCALED ONE-FOUR
\nInfo: 1-4 ELECTROSTATICS SCALED BY 1
\nInfo: MODIFIED 1-4 VDW PARAMETERS WILL BE USED
\nInfo: NO DCD TRAJECTORY OUTPUT
\nInfo: NO EXTENDED SYSTEM TRAJECTORY OUTPUT
\nInfo: NO VELOCITY DCD OUTPUT
\nInfo: NO FORCE DCD OUTPUT
\nInfo: OUTPUT FILENAME \/workspace\/examples\/stmv\/stmv-output
\nInfo: BINARY OUTPUT FILES WILL BE USED
\nInfo: NO RESTART FILE
\nInfo: SWITCHING ACTIVE
\nInfo: SWITCHING ON 10
\nInfo: SWITCHING OFF 12
\nInfo: PAIRLIST DISTANCE 13.5
\nInfo: PAIRLIST SHRINK RATE 0.01
\nInfo: PAIRLIST GROW RATE 0.01
\nInfo: PAIRLIST TRIGGER 0.3
\nInfo: PAIRLISTS PER CYCLE 2
\nInfo: PAIRLISTS ENABLED
\nInfo: MARGIN 0.48
\nInfo: HYDROGEN GROUP CUTOFF 2.5
\nInfo: PATCH DIMENSION 16.48
\nInfo: ENERGY OUTPUT STEPS 200
\nInfo: CROSSTERM ENERGY INCLUDED IN DIHEDRAL
\nInfo: TIMING OUTPUT STEPS 1
\nInfo: LANGEVIN DYNAMICS ACTIVE
\nInfo: LANGEVIN TEMPERATURE 298
\nInfo: LANGEVIN USING BBK INTEGRATOR
\nInfo: LANGEVIN DAMPING COEFFICIENT IS 5 INVERSE PS
\nInfo: LANGEVIN DYNAMICS NOT APPLIED TO HYDROGENS
\nInfo: LANGEVIN PISTON PRESSURE CONTROL ACTIVE
\nInfo: TARGET PRESSURE IS 1.01325 BAR
\nInfo: OSCILLATION PERIOD IS 200 FS
\nInfo: DECAY TIME IS 100 FS
\nInfo: PISTON TEMPERATURE IS 298 K
\nInfo: PRESSURE CONTROL IS GROUP-BASED
\nInfo: INITIAL STRAIN RATE IS 0 0 0
\nInfo: CELL FLUCTUATION IS ISOTROPIC
\nInfo: PARTICLE MESH EWALD (PME) ACTIVE
\nInfo: PME TOLERANCE 1e-06
\nInfo: PME EWALD COEFFICIENT 0.257952
\nInfo: PME INTERPOLATION ORDER 8
\nInfo: PME GRID DIMENSIONS 108 108 108
\nInfo: PME MAXIMUM GRID SPACING 2.1
\nInfo: FULL ELECTROSTATIC EVALUATION FREQUENCY 4
\nInfo: USING VERLET I (r-RESPA) MTS SCHEME.
\nInfo: C1 SPLITTING OF LONG RANGE ELECTROSTATICS
\nInfo: PLACING ATOMS IN PATCHES BY HYDROGEN GROUPS
\nInfo: RIGID BONDS TO HYDROGEN : ALL
\nInfo: ERROR TOLERANCE : 1e-08
\nInfo: MAX ITERATIONS : 100
\nInfo: RIGID WATER USING SETTLE ALGORITHM
\nInfo: NONBONDED FORCES EVALUATED EVERY 2 STEPS
\nInfo: RANDOM NUMBER SEED 3141
\nInfo: USE HYDROGEN BONDS? NO
\nInfo: STRUCTURE FILE stmv.psf.inter
\nInfo: PARAMETER file: CHARMM format!
\nInfo: PARAMETERS par_all27_prot_na.inp
\nInfo: USING ARITHMETIC MEAN TO COMBINE L-J SIGMA PARAMETERS
\nInfo: BINARY COORDINATES stmv.coor
\nInfo: SUMMARY OF PARAMETERS:
\nInfo: 250 BONDS
\nInfo: 622 ANGLES
\nInfo: 1049 DIHEDRAL
\nInfo: 73 IMPROPER
\nInfo: 0 CROSSTERM
\nInfo: 130 VDW
\nInfo: 0 VDW_PAIRS
\nInfo: 0 NBTHOLE_PAIRS
\nInfo: TIME FOR READING PSF FILE: 0.00924206
\nInfo:
\nInfo: Entering startup at 1.33416 s, 85.5117 MB of memory in use
\nInfo: Startup phase 0 took 0.000120878 s, 85.5117 MB of memory in use
\nWarning: an empty exclusion signature with index 709!
\nInfo: Startup phase 1 took 0.000654936 s, 85.5117 MB of memory in use
\nInfo: NONBONDED TABLE R-SQUARED SPACING: 0.0625
\nInfo: NONBONDED TABLE SIZE: 769 POINTS
\nInfo: INCONSISTENCY IN FAST TABLE ENERGY VS FORCE: 0.000325096 AT 11.9556
\nInfo: INCONSISTENCY IN SCOR TABLE ENERGY VS FORCE: 0.000324844 AT 11.9556
\nInfo: INCONSISTENCY IN VDWA TABLE ENERGY VS FORCE: 0.0040507 AT 0.251946
\nInfo: INCONSISTENCY IN VDWB TABLE ENERGY VS FORCE: 0.00150189 AT 0.251946
\nInfo: Running with 2 input processors.
\nInfo: Running with 1 output processors (1 of them will output simultaneously).
\nInfo: INPUT PROC LOCATIONS: 8 16
\nInfo: OUTPUT PROC LOCATIONS: 32
\nInfo: Startup phase 2 took 0.0251701 s, 241.312 MB of memory in use
\nInfo: Startup phase 3 took 0.107203 s, 244.648 MB of memory in use
\nInfo: PATCH GRID IS 13 (PERIODIC) BY 13 (PERIODIC) BY 13 (PERIODIC)
\nInfo: PATCH GRID IS 1-AWAY BY 1-AWAY BY 1-AWAY
\nInfo: LOADED 1810196 TOTAL EXCLUSIONS
\nInfo: REMOVING COM VELOCITY -0.00436736 -0.0116608 0.0017952
\nInfo: Startup phase 4 took 0.0433049 s, 560.387 MB of memory in use
\nInfo: ****************************
\nInfo: STRUCTURE SUMMARY:
\nInfo: 1066628 ATOMS
\nInfo: 769956 BONDS
\nInfo: 605872 ANGLES
\nInfo: 450875 DIHEDRALS
\nInfo: 24578 IMPROPERS
\nInfo: 0 CROSSTERMS
\nInfo: 0 EXCLUSIONS
\nInfo: 977416 RIGID BONDS
\nInfo: 2222468 DEGREES OF FREEDOM
\nInfo: 389067 HYDROGEN GROUPS
\nInfo: 4 ATOMS IN LARGEST HYDROGEN GROUP
\nInfo: 389067 MIGRATION GROUPS
\nInfo: 4 ATOMS IN LARGEST MIGRATION GROUP
\nInfo: TOTAL MASS = 6.69877e+06 amu
\nInfo: TOTAL CHARGE = 0.000168104 e
\nInfo: MASS DENSITY = 1.09115 g\/cm^3
\nInfo: ATOM DENSITY = 0.104627 atoms\/A^3
\nInfo: *****************************
\nInfo: LARGEST PATCH (1044) HAS 541 ATOMS
\nInfo: Startup phase 5 took 0.099992 s, 566.574 MB of memory in use
\nInfo: TORUS A SIZE 1 USING 0
\nInfo: TORUS B SIZE 1 USING 0
\nInfo: TORUS C SIZE 1 USING 0
\nInfo: TORUS MINIMAL MESH SIZE IS 1 BY 1 BY 1
\nInfo: Placed 100% of base nodes on same physical node as patch
\nInfo: Startup phase 6 took 0.0113301 s, 575.332 MB of memory in use
\nInfo: PME using 1 x 1 x 1 pencil grid for FFT and reciprocal sum.
\nInfo: Updated CUDA force table with 4096 elements.
\nInfo: Updated CUDA LJ table with 130 x 130 elements.
\nInfo: Updated CUDA force table with 4096 elements.
\nInfo: Updated CUDA LJ table with 130 x 130 elements.
\nInfo: Updated CUDA force table with 4096 elements.
\nInfo: Updated CUDA LJ table with 130 x 130 elements.
\nInfo: Updated CUDA force table with 4096 elements.
\nInfo: Updated CUDA LJ table with 130 x 130 elements.
\nInfo: Startup phase 7 took 9.69612 s, 1646.11 MB of memory in use
\nInfo: Startup phase 8 took 0.0201719 s, 1649.35 MB of memory in use
\nLDB: Hybrid LB being created…
\nHybridBaseLB: ThreeLevelTree is created.
\nInfo: Startup phase 9 took 0.0160291 s, 1649.5 MB of memory in use
\nInfo: CREATING 46457 COMPUTE OBJECTS
\nInfo: Found 333 unique exclusion lists needing 1076 bytes
\nInfo: Found 333 unique exclusion lists needing 1076 bytes
\nInfo: Found 333 unique exclusion lists needing 1076 bytes
\nInfo: Found 333 unique exclusion lists needing 1076 bytes
\nInfo: useSync: 0 useProxySync: 0
\nInfo: Startup phase 10 took 0.111414 s, 1686.05 MB of memory in use
\nInfo: Startup phase 11 took 0.000168085 s, 1686.05 MB of memory in use
\nInfo: Startup phase 12 took 0.000447035 s, 1686.06 MB of memory in use
\nInfo: Finished startup at 11.4663 s, 1686.06 MB of memory in use<\/p>\n

Benchmark time: 40 CPUs 0.0301711 s\/step 0.349202 days\/ns<\/strong> 2878.68 MB memory<\/p>\n

\u30c7\u30fc\u30bf\u8a73\u7d30 Oracle Linux Server 8.1 TITAN RTX x2 RTX2080Tix2 CUDA10..2 E5 2650 V4x2 namd 2.12-171025 STMV (virus) benchmark (1,066,628 atoms, periodic, PME) 0.349202 days ns<\/a><\/p>\n

GPU\u6e29\u5ea6\u63a8\u79fb Oracle Linux Server 8.1 TITAN RTX x2 RTX2080Tix2 CUDA10..2 E5 2650 V4x2 namd 2.12-171025 STMV (virus) benchmark (1,066,628 atoms, periodic, PME) 0.349202 days ns nvidia-smi<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

\u53c2\u8003\u30b5\u30a4\u30c8\uff11\u30b5\u30a4\u30c8\uff12 [chibi@oracle8 ~]$ cat \/etc\/os-release NAME=”Oracle Linux Server” VERSION=”8.1&#8 … \u7d9a\u304d\u3092\u8aad\u3080 →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[18,15],"tags":[],"class_list":["post-922","post","type-post","status-publish","format-standard","hentry","category-nvidia","category-oracle8"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/922","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=922"}],"version-history":[{"count":5,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/922\/revisions"}],"predecessor-version":[{"id":1310,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=\/wp\/v2\/posts\/922\/revisions\/1310"}],"wp:attachment":[{"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=922"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=922"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wp.study3.biz\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=922"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}