Re: [AMBER] Simulations using pmemd.cuda

From: James Starlight <jmsstarlight.gmail.com>
Date: Wed, 7 May 2014 14:39:12 +0400

Also I wounder to knmow about possible ways to monitor loading of each GPU
while performing simulations (Its strange but devise-info script found
http://ambermd.org/gpus/#Running does not allocate any GPUs ):

===================================================
Device Model Temperature Utilization
===================================================
===================================================

although both GPUs have been fully loaded according to the

own.drunk_telecaster ~/Desktop/check_CUDA $ deviceQuery
deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 2 CUDA Capable device(s)

Device 0: "GeForce GTX TITAN"
  CUDA Driver Version / Runtime Version 6.0 / 5.5
  CUDA Capability Major/Minor version number: 3.5
  Total amount of global memory: 6144 MBytes (6442123264
bytes)
  (14) Multiprocessors, (192) CUDA Cores/MP: 2688 CUDA Cores
  GPU Clock rate: 876 MHz (0.88 GHz)
  Memory Clock rate: 3004 Mhz
  Memory Bus Width: 384-bit
  L2 Cache Size: 1572864 bytes
  Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536,
65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048
layers
  Total amount of constant memory: 65536 bytes
  Total amount of shared memory per block: 49152 bytes
  Total number of registers available per block: 65536
  Warp size: 32
  Maximum number of threads per multiprocessor: 2048
  Maximum number of threads per block: 1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch: 2147483647 bytes
  Texture alignment: 512 bytes
  Concurrent copy and kernel execution: Yes with 1 copy engine(s)
  Run time limit on kernels: Yes
  Integrated GPU sharing Host Memory: No
  Support host page-locked memory mapping: Yes
  Alignment requirement for Surfaces: Yes
  Device has ECC support: Disabled
  Device supports Unified Addressing (UVA): Yes
  Device PCI Bus ID / PCI location ID: 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device
simultaneously) >

Device 1: "GeForce GTX TITAN"
  CUDA Driver Version / Runtime Version 6.0 / 5.5
  CUDA Capability Major/Minor version number: 3.5
  Total amount of global memory: 6144 MBytes (6442254336
bytes)
  (14) Multiprocessors, (192) CUDA Cores/MP: 2688 CUDA Cores
  GPU Clock rate: 876 MHz (0.88 GHz)
  Memory Clock rate: 3004 Mhz
  Memory Bus Width: 384-bit
  L2 Cache Size: 1572864 bytes
  Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536,
65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048
layers
  Total amount of constant memory: 65536 bytes
  Total amount of shared memory per block: 49152 bytes
  Total number of registers available per block: 65536
  Warp size: 32
  Maximum number of threads per multiprocessor: 2048
  Maximum number of threads per block: 1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch: 2147483647 bytes
  Texture alignment: 512 bytes
  Concurrent copy and kernel execution: Yes with 1 copy engine(s)
  Run time limit on kernels: No
  Integrated GPU sharing Host Memory: No
  Support host page-locked memory mapping: Yes
  Alignment requirement for Surfaces: Yes
  Device has ECC support: Disabled
  Device supports Unified Addressing (UVA): Yes
  Device PCI Bus ID / PCI location ID: 3 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device
simultaneously) >
> Peer access from GeForce GTX TITAN (GPU0) -> GeForce GTX TITAN (GPU1) :
Yes
> Peer access from GeForce GTX TITAN (GPU1) -> GeForce GTX TITAN (GPU0) :
Yes

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 6.0, CUDA Runtime
Version = 5.5, NumDevs = 2, Device0 = GeForce GTX TITAN, Device1 = GeForce
GTX TITAN
Result = PASS

sult = PASS
own.drunk_telecaster ~/Desktop/check_CUDA $ nvidia-smi -a
==============NVSMI LOG==============

Timestamp : Wed May 7 14:38:41 2014
Driver Version : 331.67

Attached GPUs : 2
GPU 0000:01:00.0
    Product Name : GeForce GTX TITAN
    Display Mode : N/A
    Display Active : N/A
    Persistence Mode : Disabled
    Accounting Mode : N/A
    Accounting Mode Buffer Size : N/A
    Driver Model
        Current : N/A
        Pending : N/A
    Serial Number : N/A
    GPU UUID :
GPU-eb4f78ff-f96e-2131-f72a-bad2988a3ab4
    Minor Number : 0
    VBIOS Version : 80.10.2C.00.02
    Inforom Version
        Image Version : N/A
        OEM Object : N/A
        ECC Object : N/A
        Power Management Object : N/A
    GPU Operation Mode
        Current : N/A
        Pending : N/A
    PCI
        Bus : 0x01
        Device : 0x00
        Domain : 0x0000
        Device Id : 0x100510DE
        Bus Id : 0000:01:00.0
        Sub System Id : 0x103510DE
        GPU Link Info
            PCIe Generation
                Max : N/A
                Current : N/A
            Link Width
                Max : N/A
                Current : N/A
        Bridge Chip
            Type : N/A
            Firmware : N/A
    Fan Speed : 57 %
    Performance State : N/A
    Clocks Throttle Reasons : N/A
    FB Memory Usage
        Total : 6143 MiB
        Used : 393 MiB
        Free : 5750 MiB
    BAR1 Memory Usage
        Total : N/A
        Used : N/A
        Free : N/A
    Compute Mode : Default
    Utilization
        Gpu : N/A
        Memory : N/A
    Ecc Mode
        Current : N/A
        Pending : N/A
    ECC Errors
        Volatile
            Single Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
            Double Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
        Aggregate
            Single Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
            Double Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
    Retired Pages
        Single Bit ECC : N/A
        Double Bit ECC : N/A
        Pending : N/A
    Temperature
        Gpu : 80 C
    Power Readings
        Power Management : N/A
        Power Draw : N/A
        Power Limit : N/A
        Default Power Limit : N/A
        Enforced Power Limit : N/A
        Min Power Limit : N/A
        Max Power Limit : N/A
    Clocks
        Graphics : N/A
        SM : N/A
        Memory : N/A
    Applications Clocks
        Graphics : N/A
        Memory : N/A
    Default Applications Clocks
        Graphics : N/A
        Memory : N/A
    Max Clocks
        Graphics : N/A
        SM : N/A
        Memory : N/A
    Compute Processes : N/A

GPU 0000:03:00.0
    Product Name : GeForce GTX TITAN
    Display Mode : N/A
    Display Active : N/A
    Persistence Mode : Disabled
    Accounting Mode : N/A
    Accounting Mode Buffer Size : N/A
    Driver Model
        Current : N/A
        Pending : N/A
    Serial Number : N/A
    GPU UUID :
GPU-2cb03c69-ec5d-d3ff-5d87-78f55485e053
    Minor Number : 1
    VBIOS Version : 80.10.2C.00.02
    Inforom Version
        Image Version : N/A
        OEM Object : N/A
        ECC Object : N/A
        Power Management Object : N/A
    GPU Operation Mode
        Current : N/A
        Pending : N/A
    PCI
        Bus : 0x03
        Device : 0x00
        Domain : 0x0000
        Device Id : 0x100510DE
        Bus Id : 0000:03:00.0
        Sub System Id : 0x103510DE
        GPU Link Info
            PCIe Generation
                Max : N/A
                Current : N/A
            Link Width
                Max : N/A
                Current : N/A
        Bridge Chip
            Type : N/A
            Firmware : N/A
    Fan Speed : 58 %
    Performance State : N/A
    Clocks Throttle Reasons : N/A
    FB Memory Usage
        Total : 6143 MiB
        Used : 310 MiB
        Free : 5833 MiB
    BAR1 Memory Usage
        Total : N/A
        Used : N/A
        Free : N/A
    Compute Mode : Default
    Utilization
        Gpu : N/A
        Memory : N/A
    Ecc Mode
        Current : N/A
        Pending : N/A
    ECC Errors
        Volatile
            Single Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
            Double Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
        Aggregate
            Single Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
            Double Bit
                Device Memory : N/A
                Register File : N/A
                L1 Cache : N/A
                L2 Cache : N/A
                Texture Memory : N/A
                Total : N/A
    Retired Pages
        Single Bit ECC : N/A
        Double Bit ECC : N/A
        Pending : N/A
    Temperature
        Gpu : 80 C
    Power Readings
        Power Management : N/A
        Power Draw : N/A
        Power Limit : N/A
        Default Power Limit : N/A
        Enforced Power Limit : N/A
        Min Power Limit : N/A
        Max Power Limit : N/A
    Clocks
        Graphics : N/A
        SM : N/A
        Memory : N/A
    Applications Clocks
        Graphics : N/A
        Memory : N/A
    Default Applications Clocks
        Graphics : N/A
        Memory : N/A
    Max Clocks
        Graphics : N/A
        SM : N/A
        Memory : N/A
    Compute Processes : N/A


could some one detect something unusual in these logs?

James



2014-05-07 14:18 GMT+04:00 James Starlight <jmsstarlight.gmail.com>:

> Thanks, Marc! It works perfect.
>
>
> James
>
>
> 2014-05-07 14:10 GMT+04:00 Marc van der Kamp <marcvanderkamp.gmail.com>:
>
> Hi James,
>> There are more experienced people here that can provide better informed,
>> fuller answers, but I hope this will help:
>>
>> >
>> >
>> > 1) How I could use my second GPU for second system?
>> >
>>
>> Set the environment variable *CUDA_VISIBLE_DEVICES *(you can search for
>> this in the Amber mailing list to see previous posts).
>> *CUDA_VISIBLE_DEVICES=0* should point to your first GPU, and if you then
>> set *CUDA_VISIBLE_DEVICES=1 *you should use your other one.
>>
>>
>> > 2)Does it possible to use 3 CPU cores for each GPU obtaining increase in
>> > performance? Should I use MP for such tasks?
>> >
>> > The pmemd.cuda code runs completely on the GPU, using CPU only for i/o.
>> So
>> I'm pretty sure there will not be any increase in performance when using
>> more CPU cores.
>>
>>
>> > Thanks for help,
>> >
>> > James
>> > _______
>>
>>
>> Hope this helps,
>> Marc
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed May 07 2014 - 04:00:02 PDT
Custom Search